WO2021147368A1 - 调整服务等级的方法、装置、设备、系统及存储介质 - Google Patents

调整服务等级的方法、装置、设备、系统及存储介质 Download PDF

Info

Publication number
WO2021147368A1
WO2021147368A1 PCT/CN2020/118590 CN2020118590W WO2021147368A1 WO 2021147368 A1 WO2021147368 A1 WO 2021147368A1 CN 2020118590 W CN2020118590 W CN 2020118590W WO 2021147368 A1 WO2021147368 A1 WO 2021147368A1
Authority
WO
WIPO (PCT)
Prior art keywords
service level
target service
queue
data
parameters
Prior art date
Application number
PCT/CN2020/118590
Other languages
English (en)
French (fr)
Inventor
陈爽
李磊
龚钧
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to KR1020227028085A priority Critical patent/KR102693696B1/ko
Priority to JP2022544401A priority patent/JP7531594B2/ja
Priority to EP20915290.9A priority patent/EP4084409A4/en
Publication of WO2021147368A1 publication Critical patent/WO2021147368A1/zh
Priority to US17/814,071 priority patent/US20220407808A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
    • H04L41/5003—Managing SLA; Interaction between SLA and QoS
    • H04L41/5019—Ensuring fulfilment of SLA
    • H04L41/5025—Ensuring fulfilment of SLA by proactively reacting to service quality change, e.g. by reconfiguration after service quality degradation or upgrade
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
    • H04L41/5003—Managing SLA; Interaction between SLA and QoS
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0876—Network utilisation, e.g. volume of load or congestion level
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2425—Traffic characterised by specific attributes, e.g. priority or QoS for supporting services specification, e.g. SLA
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/30—Flow control; Congestion control in combination with information about buffer occupancy at either end or at transit nodes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/52—Queue scheduling by attributing bandwidth to queues
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/56—Queue scheduling implementing delay-aware scheduling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0813—Configuration setting characterised by the conditions triggering a change of settings
    • H04L41/082—Configuration setting characterised by the conditions triggering a change of settings the condition being updates or upgrades of network functionality
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/16—Threshold monitoring

Definitions

  • This application relates to the field of communications, in particular to methods, devices, equipment, systems, and storage media for adjusting service levels.
  • 5G fifth-generation
  • uRLLC ultra-reliable and low-latency communication
  • IP Internet Protocol
  • SLA Internet Protocol
  • This application provides a method, device, equipment, system, and storage medium for adjusting service levels, so as to improve the utilization of network resources on the premise of ensuring the end-to-end delay requirements of the service.
  • a method for adjusting a service level includes: a control device obtains at least one queue status information, remaining data flow parameters, current data flow parameters, and abnormal information reporting of a target service level of a first network device At least one type of related information in the number of times.
  • the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level based on that any information in the related information does not meet the threshold corresponding to any information.
  • the at least one queue status information of the target service level includes one or more queue status information of the target service level.
  • each service level is bound to one queue or multiple queues of the first network device port, where the multiple queues are referred to as a set of queues.
  • the queue status information of the target service level refers to the queue status information of one or more queues in the group of queues bound to the target service level, or,
  • the queue status information of the target service level refers to the overall queue status information of a group of queues bound to the target service level.
  • the queue status information includes queue status information including but not limited to one or more of buffer occupation, queuing delay, and packet count of the local queue.
  • the buffer occupancy of the queue indicates the size of the buffer occupied by the messages in the queue bound to the target service level;
  • the message queuing delay indicates the queuing delay of the messages in the queue bound to the target service level, including the message entering the buffer The time interval until the message is scheduled;
  • the message count indicates the amount of data per unit time or the number of packets in the queue bound to the target service level.
  • the remaining data flow parameter of the target service level includes the flow size of the remaining allowable data flow of the target service level of the first network device.
  • the abnormal information of the target service level includes at least one of abnormal information of queue buffer occupancy, abnormal information of queuing delay, abnormal information of packet count, and error information.
  • the current data flow parameter of the target service level includes the current data flow size of the target service level of the first network device.
  • the error information is ERROR information sent by the network device to the control device when the traffic of the target service level exceeds the admission restriction or the resource threshold.
  • the utilization rate of network resources is further improved on the premise of ensuring the end-to-end delay of the business.
  • the relevant information includes at least one queue status information of the target service level, and based on the fact that any queue status information in the at least one queue status information is less than the lower threshold corresponding to any queue status information, the control device is based on The maximum delay associated with the target service level adjusts the parameters of the target service level.
  • the control device is triggered to adjust the target service level according to the maximum delay associated with the target service level.
  • the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level, so that when the queue status information changes, the end-to-end delay of the service can be guaranteed , Improve the utilization of network resources.
  • the relevant information includes the current data flow parameters of the target service level, the current data flow parameters based on the target service level are less than the lower threshold corresponding to the current data flow parameters, and the control device is based on the current data flow parameters associated with the target service level.
  • the maximum delay adjusts the parameters of the target service level.
  • the control device adjusts the target service level parameter according to the maximum delay associated with the target service level, so that the end-to-end delay of the service can be guaranteed when the current traffic change is considered , Improve the utilization of network resources.
  • the relevant information includes the remaining data flow parameters of the target service level, the remaining data flow parameters based on the target service level exceed the upper threshold value corresponding to the remaining data flow parameters, and the control device is based on the remaining data flow parameters associated with the target service level.
  • the maximum delay adjusts the parameters of the target service level.
  • the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level, so that the end-to-end delay of the service can be guaranteed when the remaining traffic changes are considered , And further improve the utilization of network resources.
  • the parameters of the target service level include queue resource parameters and data flow constraint parameters.
  • the control device determines the first update value of the data flow constraint parameter, and when the maximum time delay remains unchanged, determines the first update value of the queue resource parameter according to the first update value of the data flow constraint parameter.
  • the control device adjusts the data flow constraint parameter according to the first update value of the data flow constraint parameter, and adjusts the queue resource parameter according to the first update value of the queue resource parameter.
  • the first update value of the queue resource parameter includes the first update value of the queue bandwidth and the first update value of the queue buffer.
  • the first update value of the data flow constraint parameter includes the first update value of the burst volume threshold of the data flow and the first update value of the average rate threshold of the data flow.
  • the control device determines the first update value of the queue bandwidth.
  • the maximum amount of data that the control device can forward before the parameter adjustment of the target service level according to the multiple service levels, the maximum message length of the queues corresponding to the multiple service levels, the first update value of the burst threshold of the data flow, and the data flow determines the first update value of the queue buffer.
  • the first maximum amount of data that can be forwarded after the parameter adjustment of the target service level at the target service level, and the parameter adjustment of multiple service levels at the target service level is determined according to the following formula
  • the maximum packet length of the queues corresponding to the multiple service levels, and the burst threshold of the data flow is determined according to the following formula
  • the above C is the bandwidth of the port corresponding to the target service level, Is the first maximum data volume that can be forwarded by the target service level after the parameters of the target service level are adjusted, and Q j is the maximum data volume that can be forwarded by the j-th service level, Is the maximum amount of data that can be forwarded by the target service level before the parameters of the target service level are adjusted, It is the maximum amount of data that the non-delay guarantee queue can forward before the parameters of the target service level are adjusted; L max, L is the maximum packet length in the low priority queue, and L n+1 is the maximum data in the non-delay guarantee queue Packet length, L j is the maximum packet length in the j-th service level queue, Is the first updated value of the burst threshold of the data stream, The burst threshold of the data stream before the parameter adjustment of the target service level, ⁇ b i1 is the decrease value of the burst threshold of the data stream of the target service level, Is the first updated value of the average rate
  • the related information includes at least one queue status information of the target service level, and based on the fact that any queue status information in the at least one queue status information exceeds the upper threshold corresponding to any queue status information, the control device is based on The maximum delay associated with the target service level adjusts the parameters of the target service level.
  • the control device takes the queue status information including the buffer occupancy of the local queue, packet queuing delay, and packet count as an example, if it is detected that the buffer occupancy of the local queue of the target service level exceeds the upper limit of the buffer occupancy threshold, or the packet queuing delay If the upper threshold of the message queuing delay is exceeded, or the message count exceeds the upper threshold of the message count, the control device is triggered to adjust the parameters of the target service level according to the maximum delay associated with the target service level.
  • the relevant information includes the current data flow parameters of the target service level, the current data flow parameters based on the target service level exceed the upper threshold value corresponding to the current data flow parameters, and the control device is based on the current data flow parameters associated with the target service level.
  • the maximum delay adjusts the parameters of the target service level.
  • the related information includes the remaining data flow parameters of the target service level, and the remaining data flow parameters based on the target service level are less than the lower threshold corresponding to the remaining data flow parameters, and the control device is based on the remaining data flow parameters associated with the target service level.
  • the maximum delay adjusts the parameters of the target service level.
  • the relevant information includes the number of reports of abnormal information of the target service level, and the number of reports of abnormal information based on the target service level exceeds the upper limit of the threshold corresponding to the number of reports of abnormal information.
  • the maximum delay associated with the level adjusts the parameters of the target service level.
  • the first network device reports abnormal information to the control device based on the updated values of the configuration parameters of the target service level that the local resources do not meet the target service level when the maximum delay is unchanged, and the abnormal information refers to the abnormal service information of the first network device.
  • the abnormal service quality information includes at least one of queue buffer occupation that exceeds a threshold, message queuing delay, message count information, and error information.
  • the parameters of the target service level include queue resource parameters and data flow constraint parameters.
  • the control device determines the second updated value of the data flow constraint parameter. When the maximum delay is unchanged, the control device determines the second update value of the queue resource parameter according to the second update value of the data flow constraint parameter. Based on the second update value of the queue resource parameter being less than or equal to the resource threshold, the control device adjusts the data flow restriction parameter according to the second update value of the data flow restriction parameter, and adjusts the queue resource parameter according to the second update value of the queue resource parameter.
  • the configuration parameters of the target service level include queue resource parameters and data flow constraint parameters.
  • the control device determines the second updated value of the data flow constraint parameter.
  • the control device determines the second update value of the queue resource parameter according to the second update value of the data flow constraint parameter.
  • the control device switches the target data stream corresponding to the target service level to another service level, and the other service levels include other service levels of the first network device or service levels of other devices.
  • the configuration parameters of the target service level include queue resource parameters and data flow constraint parameters.
  • the control device determines the second updated value of the data flow constraint parameter. When the maximum delay is unchanged, the control device determines the second update value of the queue resource parameter according to the second update value of the data flow constraint parameter. Based on the second update value of the queue resource parameter being greater than the resource threshold, the control device determines the third update value of the queue resource parameter according to the second update value of the data flow constraint parameter, and determines the update value of the parameter of the service level that can preempt the resource. The updated value of the parameter of the service level of the preemptible resource is used to make the third updated value of the queue resource parameter meet the constraint condition.
  • the control device adjusts the data flow constraint parameter according to the second update value of the data flow constraint parameter, adjusts the queue resource parameter according to the third update value of the queue resource parameter, and adjusts the resource preemptible service according to the updated value of the parameter of the service level of the resource preemptible The parameters of the level.
  • the second update value of the queue resource parameter includes the second update value of the queue bandwidth and the second update value of the queue buffer
  • the second update value of the data flow constraint parameter includes the burst amount of the data flow.
  • the second update value of the queue resource parameter is determined according to the second update value of the data flow constraint parameter, including: under the condition that the maximum delay is unchanged, determining the port corresponding to the target service level Bandwidth, the second maximum amount of data that can be forwarded by the target service level after the parameters of the target service level are adjusted, the maximum amount of data that can be forwarded by the multiple service levels of the first network device before the parameters of the target service level are adjusted, and multiple service levels
  • the maximum message length of the corresponding queue according to the bandwidth of the port corresponding to the target service level, the second maximum amount of data that can be forwarded after the parameter adjustment of the target service level at the target service level, and the parameter adjustment of multiple service levels at the target service level
  • the maximum amount of data that can be forwarded before determines the second update value of the queue bandwidth; the maximum amount of data that can be forwarded before the parameter adjustment of the target service level according to multiple service levels and the maximum packet length of the queue corresponding to multiple service levels,
  • the second maximum amount of data that can be forwarded after the parameter adjustment of the target service level at the target service level, and the parameter adjustment of multiple service levels at the target service level The maximum amount of data that can be forwarded before determining the second update value of the queue bandwidth, including: according to the bandwidth of the port corresponding to the target service level, the second maximum amount of data that can be forwarded after the target service level is adjusted after the parameters of the target service level are adjusted, and each The maximum amount of data that can be forwarded by the service level before the parameters of the target service level are adjusted, and the second update value of the queue bandwidth is determined according to the following formula
  • the maximum packet length of the queues corresponding to the multiple service levels, and the burst threshold of the data flow determine the second update value of the queue buffer, including: the maximum amount of data that can be forwarded before the parameter adjustment of the target service level according to multiple service levels and multiple The maximum message length of the queue corresponding to the service level, the second update value of the burst threshold of the data flow, and the second update value of the average rate threshold of the data flow, determine the second update value of the queue buffer according to the following formula
  • C is the bandwidth of the port corresponding to the target service level
  • Q j is the maximum data volume that can be forwarded by the j-th service level
  • L max L is the maximum packet length in the low priority queue
  • L n+1 is the maximum amount in the non-delay guarantee queue
  • Packet length L j is the maximum packet length in the j-th service level queue
  • ⁇ b i2 is the increase value of the burst threshold of the data stream
  • the second update value of the data flow restriction parameter includes the second update value of the burst threshold value of the data flow and the second update value of the average rate threshold of the data flow; according to the data flow restriction parameter
  • the second update value determines the third update value of the queue resource parameter, including: determining the bandwidth of the port corresponding to the target service level, the third maximum amount of data that the target service level can forward after the parameters of the target service level are adjusted, and the first The maximum amount of data that can be forwarded by the multiple service levels of the network equipment before the parameter adjustment of the target service level and the maximum message length of the queue corresponding to the multiple service levels; according to the bandwidth of the port corresponding to the target service level, the target service level is The third maximum amount of data that can be forwarded after the parameters of the target service level are adjusted and the maximum amount of data that can be forwarded by multiple service levels before the parameters of the target service level are adjusted to determine the third update value of the queue bandwidth; according to the multiple service levels The maximum amount of data that can be forwarded before the parameter
  • Determine the updated value of the parameters of the service level of the preemptible resource including: according to the bandwidth of the port corresponding to the target service level, the maximum amount of data that can be forwarded after the adjustment of the service level of the preemptable resource, and the parameters of multiple service levels at the target service level
  • the maximum amount of data that can be forwarded before adjustment determine the updated value of the queue bandwidth of the service level that can grab resources; the maximum amount of data that can be forwarded before the parameter adjustment of the target service level according to multiple service levels, and the queues corresponding to multiple service levels
  • the updated value of determines the updated value of the queue cache of the service level that can preempt resources.
  • the third maximum amount of data that can be forwarded after the target service level is adjusted after the parameters of the target service level are adjusted, and the number of service levels within the target service level
  • the maximum amount of data that can be forwarded before the parameter adjustment, and the third update value of the queue bandwidth is determined, including: according to the bandwidth of the port corresponding to the target service level, the third largest target service level that can be forwarded after the parameter adjustment of the target service level
  • the amount of data and the maximum amount of data that can be forwarded by multiple service levels before the parameters of the target service level are adjusted, and the third update value of the queue bandwidth is determined according to the following formula
  • the maximum amount of data that can be forwarded after the adjustment of the service level that can preempt resources determines the service that can preempt resources Level of queue bandwidth
  • the maximum packet length of the queues corresponding to the multiple service levels, and the burst threshold of the data flow determine the third update value of the queue buffer, including: the maximum amount of data that can be forwarded before the parameter adjustment of the target service level according to multiple service levels and multiple The maximum message length of the queue corresponding to the service level, the maximum amount of data that can be forwarded after the adjustment of the service level that can preempt resources, the maximum message length, the second update value of the burst threshold of the data flow, and the average rate threshold of the data flow
  • the second update value of, the third update value of the queue buffer is determined according to the following formula
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level the maximum packet length of the queues corresponding to multiple service levels, the maximum amount of data that can be forwarded after the adjustment of the service level that can preempt resources, and the maximum report
  • the length of the text, the update value of the burst threshold value of the data flow of the service level that can preempt resources, and the update value of the average rate threshold of the data flow determine the queue cache of the service level of the resource that can be preempted
  • C is the bandwidth of the port corresponding to the target service level
  • Q j is the maximum amount of data that can be forwarded by the jth service level
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level Is the maximum amount of data that the non-delay guarantee queue can forward before the parameters of the target service level are adjusted
  • L max L is the maximum packet length in the low priority queue
  • L n+1 is the maximum amount in the non-delay guarantee queue
  • L j is the maximum packet length in the j-th service level queue
  • ⁇ b i2 is the increase value of the burst threshold of the data stream
  • the method further includes: the control device sends the updated value of the adjusted target service level parameter to the first network device, and instructs the first network device to follow The updated value of the parameter of the target service level adjusts the parameter of the target service level.
  • control device obtains the related information of the target service level of the first network device includes: the control device receives the related information of the target service level sent by the first network device within a preset period.
  • a device for adjusting service level includes:
  • the obtaining module is used to obtain relevant information of the target service level of the first network device, the relevant information including at least one queue status information of the target service level, remaining data flow parameters, current data flow parameters, and the number of times of reporting abnormal information At least one type of information.
  • the adjustment module is configured to adjust the parameters of the target service level according to the maximum delay associated with the target service level based on that any information in the related information does not meet the threshold corresponding to the any information.
  • the related information includes at least one queue status information of the target service level.
  • the adjustment module is configured to adjust the maximum delay according to the maximum delay associated with the target service level based on that any queue state information in the at least one queue state information is less than a lower threshold corresponding to the any queue state information The parameters of the target service level.
  • the related information includes current data flow parameters of the target service level.
  • the adjustment module is configured to adjust the parameters of the target service level according to the maximum delay associated with the target service level based on that the current data flow parameter of the target service level is less than the lower threshold corresponding to the current data flow parameter .
  • the related information includes remaining data stream parameters of the target service level.
  • the adjustment module is configured to adjust the parameters of the target service level based on the maximum delay associated with the target service level based on the remaining data flow parameters of the target service level exceeding the upper threshold value corresponding to the remaining data flow parameters .
  • the parameters of the target service level include queue resource parameters and data flow restriction parameters.
  • the adjustment module is configured to determine the first update value of the data flow constraint parameter; under the condition that the maximum delay remains unchanged, determine the value of the queue resource parameter according to the first update value of the data flow constraint parameter A first update value; adjust the data flow restriction parameter according to the first update value of the data flow restriction parameter, and adjust the queue resource parameter according to the first update value of the queue resource parameter.
  • the first update value of the queue resource parameter includes the first update value of the queue bandwidth and the first update value of the queue buffer
  • the first update value of the data flow constraint parameter includes the data flow The first update value of the burst threshold and the first update value of the average rate threshold of the data stream
  • the adjustment module is configured to determine the bandwidth of the port corresponding to the target service level, and the first target service level that can be forwarded after the parameters of the target service level are adjusted under the condition that the maximum delay is unchanged.
  • the maximum amount of data, the maximum amount of data that can be forwarded by the multiple service levels of the first network device before the parameters of the target service level are adjusted, and the maximum packet length of the queues corresponding to the multiple service levels;
  • the first maximum amount of data that can be forwarded after the target service level parameters are adjusted, and the multiple service levels before the target service level parameters are adjusted.
  • the maximum amount of data that can be forwarded determines the first update value of the queue bandwidth.
  • the maximum amount of data that can be forwarded before the parameters of the target service level are adjusted, the maximum packet length of the queues corresponding to the multiple service levels, and the first threshold of the burst volume threshold of the data flow
  • An update value and the first update value of the average rate threshold of the data stream determine the first update value of the queue buffer.
  • the adjustment module is configured to adjust according to the bandwidth of the port corresponding to the target service level, the first maximum amount of data that can be forwarded by the target service level after the parameters of the target service level are adjusted, and The maximum amount of data that can be forwarded by the multiple service levels before the parameters of the target service level are adjusted, and the first update value of the queue bandwidth is determined according to the following formula
  • the adjustment module is configured to adjust the maximum amount of data that can be forwarded according to the multiple service levels before the parameters of the target service level are adjusted and the number of queues corresponding to the multiple service levels.
  • the maximum packet length, the first update value of the burst threshold value of the data flow, and the first update value of the average rate threshold of the data flow determine the first update value of the queue buffer according to the following formula
  • the related information includes at least one queue status information of the target service level; the adjustment module is configured to exceed all queue status information based on the at least one queue status information in the at least one queue status information.
  • the upper threshold value corresponding to any of the queue status information adjusts the parameters of the target service level according to the maximum delay associated with the target service level.
  • the related information includes current data flow parameters of the target service level
  • the adjustment module is configured to adjust the parameters of the target service level according to the maximum delay associated with the target service level based on the current data flow parameter of the target service level exceeding the upper threshold value corresponding to the current data flow parameter .
  • the related information includes remaining data stream parameters of the target service level.
  • the adjustment module is configured to adjust the parameters of the target service level according to the maximum delay associated with the target service level based on that the remaining data flow parameters of the target service level are less than the lower threshold corresponding to the remaining data flow parameters .
  • the related information includes the number of times of reporting abnormal information of the target service level.
  • the adjustment module is configured to adjust the target service level based on the maximum time delay associated with the target service level based on the number of reports of abnormal information of the target service level exceeding the upper threshold value corresponding to the number of reports of the abnormal information Parameters.
  • the parameters of the target service level include a queue resource parameter and a data flow restriction parameter; the adjustment module is used to determine the second update value of the data flow restriction parameter.
  • the data flow restriction parameter is adjusted according to the second update value of the data flow restriction parameter, and the data flow restriction parameter is adjusted according to the second update value of the queue resource parameter. Queue resource parameters.
  • the configuration parameters of the target service level include queue resource parameters and data flow restriction parameters.
  • the adjustment module is used to determine the second update value of the data flow constraint parameter.
  • the target data stream corresponding to the target service level is switched to another service level, where the other service level includes other service levels of the first network device or other service levels The service level of the device.
  • the configuration parameters of the target service level include queue resource parameters and data flow restriction parameters.
  • the adjustment module is configured to determine the second update value of the data flow constraint parameter; in the case that the maximum delay remains unchanged, determine the value of the queue resource parameter according to the second update value of the data flow constraint parameter The second update value.
  • the third update value of the queue resource parameter is determined according to the second update value of the data flow constraint parameter, and the update of the parameter of the service level that can preempt the resource is determined Value, the updated value of the parameter of the service level of the preemptible resource is used to make the third updated value of the queue resource parameter meet the constraint condition.
  • Adjust the data flow restriction parameter according to the second update value of the data flow restriction parameter adjust the queue resource parameter according to the third update value of the queue resource parameter, and adjust the queue resource parameter according to the parameter of the service level of the preemptible resource
  • the updated value adjusts the parameter of the service level of the preemptible resource.
  • the second update value of the queue resource parameter includes a second update value of the queue bandwidth and a second update value of the queue buffer
  • the second update value of the data flow constraint parameter includes a data flow The second update value of the burst threshold and the second update value of the average rate threshold of the data stream.
  • the adjustment module is configured to determine the bandwidth of the port corresponding to the target service level and the second maximum that the target service level can be forwarded after the parameters of the target service level are adjusted under the condition that the maximum delay is unchanged.
  • the amount of data, the maximum amount of data that can be forwarded by the multiple service levels of the first network device before the parameters of the target service level are adjusted, and the maximum packet length of the queues corresponding to the multiple service levels.
  • the second maximum amount of data that can be forwarded after the target service level parameters are adjusted, and the multiple service levels before the target service level parameters are adjusted.
  • the maximum amount of data that can be forwarded determines the second updated value of the queue bandwidth.
  • the maximum amount of data that can be forwarded before the parameters of the target service level are adjusted, the maximum packet length of the queues corresponding to the multiple service levels, and the first threshold of the burst volume threshold of the data flow
  • the second update value and the second update value of the average rate threshold of the data stream determine the second update value of the queue buffer.
  • the adjustment module is configured to adjust according to the bandwidth of the port corresponding to the target service level, the second maximum amount of data that can be forwarded by the target service level after the parameters of the target service level are adjusted, and The maximum amount of data that can be forwarded by each service level before the parameters of the target service level are adjusted, and the second update value of the queue bandwidth is determined according to the following formula
  • the adjustment module is configured to adjust the maximum amount of data that can be forwarded according to the multiple service levels before the parameters of the target service level are adjusted and the number of queues corresponding to the multiple service levels.
  • the maximum packet length, the second update value of the burst threshold of the data flow, and the second update value of the average rate threshold of the data flow determine the second update value of the queue buffer according to the following formula
  • the second update value of the data flow restriction parameter includes the second update value of the burst threshold value of the data flow and the second update value of the average rate threshold of the data flow.
  • the adjustment module is configured to determine the bandwidth of the port corresponding to the target service level, the third maximum amount of data that the target service level can forward after the parameters of the target service level are adjusted, and the amount of data of the first network device. The maximum amount of data that can be forwarded for each service level before the parameters of the target service level are adjusted and the maximum message length of the queues corresponding to the multiple service levels.
  • the third maximum amount of data that can be forwarded by the target service level after the parameters of the target service level are adjusted, and the parameters of the multiple service levels at the target service level
  • the maximum amount of data that can be forwarded before adjustment is determined, and the third updated value of the queue bandwidth is determined.
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level, the maximum message length of the queues corresponding to the multiple service levels, and the service level of the preemptible resource can be adjusted after adjustment.
  • the adjustment module is configured to adjust the maximum amount of data that can be forwarded according to the bandwidth of the port corresponding to the target service level, the service level of the preemptible resource, and the parameters of the multiple service levels at the target service level. The maximum amount of data that can be forwarded before adjustment is determined, and the updated value of the queue bandwidth of the service level that can preempt the resource is determined.
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level, the maximum message length of the queues corresponding to the multiple service levels, and the service level of the preemptible resource can be adjusted after adjustment.
  • the adjustment module is configured to adjust the third largest data that can be forwarded according to the bandwidth of the port corresponding to the target service level and the target service level after the parameters of the target service level are adjusted And the maximum amount of data that the multiple service levels can forward before the parameters of the target service level are adjusted, the third update value of the queue bandwidth is determined according to the following formula
  • the maximum amount of data that can be forwarded after the adjustment of the service level that can preempt resources, and the maximum data that can be forwarded by the multiple service levels before the parameters of the target service level are adjusted Quantity, determine the queue bandwidth of the service level that can preempt the resource
  • the adjustment module is configured to adjust the maximum amount of data that can be forwarded according to the multiple service levels before the parameters of the target service level are adjusted and the number of queues corresponding to the multiple service levels.
  • the maximum message length, the maximum amount of data that can be forwarded after the adjustment of the service level of the preemptible resource and the maximum message length, the second update value of the burst threshold of the data flow, and the average rate threshold of the data flow is determined according to the following formula
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level, the maximum message length of the queues corresponding to the multiple service levels, and the service level of the preemptible resources can be adjusted
  • the device further includes:
  • the sending module is configured to send the adjusted update value of the parameter of the target service level to the first network device, and instruct the first network device to adjust the target service level according to the updated value of the parameter of the target service level Parameters.
  • the acquiring module is configured to receive the target service level related information sent by the first network device within a preset period.
  • a network device that executes the method in the first aspect or any implementation manner of the first aspect.
  • the network device includes a unit for executing the method in the implementation manner of the first aspect.
  • a computer-readable storage medium includes a computer program or instruction.
  • the computer program or instruction When the computer program or instruction is executed on a computer, the computer executes the first aspect and any one of the above. The method described.
  • a network device in a fifth aspect, includes a memory and a processor, and a computer program or at least one instruction is stored in the memory, and the computer program or at least one instruction is loaded and executed by the processor, In order to achieve the above first aspect and any of the methods described above.
  • a communication device which includes a transceiver, a memory, and a processor.
  • the transceiver, the memory, and the processor communicate with each other through an internal connection path
  • the memory is used to store computer programs or instructions
  • the processor is used to execute the computer programs or instructions stored in the memory to control the transceiver to receive signals, And control the transceiver to send a signal
  • the processor executes the instruction stored in the memory, the processor is caused to execute the method in the first aspect or any one of the possible implementation manners of the first aspect.
  • processors there are one or more processors, and one or more memories.
  • the memory may be integrated with the processor, or the memory and the processor may be provided separately.
  • the memory can be a non-transitory (non-transitory) memory, such as a read only memory (ROM), which can be integrated with the processor on the same chip, or can be set in different On the chip, the embodiment of the present application does not limit the type of the memory and the setting mode of the memory and the processor.
  • ROM read only memory
  • a computer program includes: computer program code, which when the computer program code is run by a computer, causes the computer to execute the first aspect or the first aspect described above.
  • the method in any one of the possible implementations of the aspect.
  • a chip including a processor, configured to call and run computer programs or instructions stored in the memory from a memory, so that a communication device installed with the chip executes the first aspect or the first aspect described above.
  • the method in any one of the possible implementations of the aspect.
  • another chip including: an input interface, an output interface, a processor, and a memory.
  • the input interface, output interface, the processor, and the memory are connected by an internal connection path, and the processing
  • the processor is configured to execute the code in the memory, and when the code is executed, the processor is configured to execute the foregoing first aspect or the method in any one of the possible implementation manners of the first aspect.
  • a system for adjusting service levels including a control device and a first network device in any possible implementation manner of the second aspect or the second aspect, where the first network device is used to send a target to the control device
  • the related information of the service level, the related information of the target service level includes at least one of at least one queue status information, remaining data flow parameters, current data flow parameters, and the number of times of reporting abnormal information of the target service level.
  • FIG. 1 is a schematic diagram of a service level adjustment scenario provided by an embodiment of the application
  • FIG. 2 is a schematic diagram of a scenario for adjusting a service level provided by an embodiment of the application
  • FIG. 3 is a schematic diagram of a system architecture for adjusting a service level according to an embodiment of the application
  • FIG. 4 is a schematic flowchart of a method for adjusting a service level provided by an embodiment of this application
  • FIG. 5 is a schematic flowchart of a method for adjusting a service level provided by an embodiment of this application
  • FIG. 6 is a schematic diagram of an egress port queue for adjusting the service level according to an embodiment of the application
  • FIG. 7 is a schematic diagram of network calculation for adjusting service level according to an embodiment of the application.
  • FIG. 8 is a schematic structural diagram of an apparatus for adjusting a service level provided by an embodiment of this application.
  • FIG. 9 is a schematic structural diagram of a network device provided by an embodiment of this application.
  • FIG. 10 is a schematic structural diagram of a network device provided by an embodiment of this application.
  • FIG. 11 is a schematic structural diagram of a device for adjusting a service level provided by an embodiment of this application.
  • the traditional network quality of service (QoS) guarantee includes the integrated service (IntServ) model and the differentiated service (Differentiated Services, DiffServ) model.
  • IntServ integrated service
  • DiffServ Differentiated Services
  • the IntServ model relies on network devices in the network for flow-by-flow state maintenance, resource reservation, and admission control.
  • the DiffServ model does not require network equipment to maintain flow-by-flow of service data, but instead implements a priority-based hop-by-hop behavior forwarding operation on the message according to the differentiated service type carried in the message.
  • the IntServ model can guarantee the end-to-end QoS of the service, but its scalability is poor and it is difficult to deploy on a large scale.
  • the DiffServ model has good scalability, it is limited by the relative characteristics of the priority strategy of network equipment, and cannot provide a definite forwarding capability, and thus cannot strictly guarantee the end-to-end delay requirements of the service data flow. Therefore, neither model can realize the end-to-end delay guarantee in the current large-scale IP network.
  • the current QoS technology based on SLA to ensure end-to-end delay does not consider the dynamic QoS adjustment technology of the service level when business requirements, real-time data flow, and network status change. If you keep the static QoS settings unchanged, there is no guarantee that they will still be optimal when the relevant conditions change.
  • the embodiment of the present application proposes a method for adjusting the service level, which can dynamically update the queue according to changes in service deployment conditions (for example, current data flow parameters and remaining data flow parameters), queue status information, and reporting times of abnormal information, etc.
  • the parameter configuration of service levels such as resource allocation and data flow admission constraints, as well as the dynamic switching of the service levels through which data flows pass, further improve the utilization of network resources while ensuring the end-to-end delay requirements of the business.
  • the following uses the scenario diagram shown in FIG. 1 as an example to introduce the method of adjusting the service level.
  • the network devices are a first network device and a second network device.
  • the first network device and the second network device maintain one or more outgoing ports for sending service data streams.
  • a service level and allocate the corresponding bandwidth, cache and other resources to the queue bound by the service level.
  • the first network device and the second network device are network devices on a path for transmitting data streams with delay requirements.
  • the first network device and the second network device respectively report one or more of their service level and associated maximum delay, confidence coefficient, data flow admission restriction and other information to the control device.
  • the controller obtains the service parameters promised by the service data flow contract, including one or more of the information such as the maximum allowable burst and the maximum allowable average rate.
  • the control device performs service deployment based on the acquired information, including selecting the forwarding path and selecting the service level to be entered by the network device at each hop on the path to ensure that the end-to-end delay requirements of the service data flow are met, and the selected service level meets Respective data flow admission constraints.
  • the control device delivers information such as the forwarding path and service level selection per hop to the network device, so that the service data flow is forwarded according to the specified path and service level.
  • the control device can be an independent physical device, as shown in Figure 1, that is physically independent of the first network device and the second network device; it can also be a functional unit deployed on the first network device or the second network device . As long as the control device has logically corresponding management and control functions, this application does not limit the existence of the control device.
  • the first network device and the second network device can be in the form of hardware or a combination of software and hardware, and are independent network devices, such as switches, routers, and other network devices with forwarding functions that are used to receive and send data streams in the network .
  • the first network device and the second network device may also be in the form of software, which is a functional module or a combination of multiple functional modules on other network devices in the network, and can be selected and designed according to specific scenario requirements.
  • the network scenario includes a controller 101 and a number of network devices, where the number of network devices includes a first network device 102, a second network device 103, and a third network device 104, etc. .
  • the first network device 102 is one or more routers or switches with a packet forwarding function.
  • the control device 101 may be in the form of a server with functions such as path calculation and resource allocation, network equipment or other software, and a combination of software and hardware.
  • the embodiment of the present application only takes the control device 101 as a controller as an example.
  • one or more service levels are set on the port through which the first network device 102 forwards the data stream, and each service level provides a corresponding forwarding service capability.
  • the deterministic forwarding service capability provided by the first network device 102 is defined as the service level of the first network device 102.
  • Each service level is bound to a queue or a group of queues in the first network device 102.
  • Each service level is associated with a delay threshold, such as the maximum delay.
  • the delay threshold indicates the allowable delay of the data flow input to the queue bound to the service level during the forwarding process of the first network device 102, and the delay threshold includes the queuing delay, processing delay and sending delay of the first network device.
  • One or more of the delays such as time delay.
  • the queuing delay refers to the time it takes for a message to enter the queue.
  • queuing delay refers to the time it takes for packets to enter the queue of the downstream traffic manager.
  • the transmission delay refers to the time required for the network device to send a message, that is, the time from sending the first bit of the message to the completion of the last bit of the message.
  • Processing delay refers to the time consumed by operations such as packet header parsing, error checking, and routing lookup when the network device receives a packet.
  • the delay threshold is the queuing delay of the first network device; in another example, the delay threshold is the sum of the queuing delay, processing delay, and sending delay of the first network device.
  • each service level may also be associated with a confidence coefficient, or reliability probability, which represents the probability that the time delay generated when the data stream passes through the service level does not exceed the above-mentioned delay threshold.
  • the confidence factor can be set to 100% or 1, which means that the generated delay does not strictly exceed the delay threshold.
  • resources need to be allocated to the queue bound to the service level, including but not limited to queue bandwidth, queue buffering, etc.
  • queue bandwidth is the ratio of the queue weight to the total weight multiplied by the bandwidth of the egress port.
  • constraints on the data flow of the input queue including but not limited to the burst threshold of the data flow entering the queue, the average rate of the data flow threshold, and so on.
  • the burst threshold of the data stream is expressed as the maximum amount of data that the data stream is allowed to reach within a set time
  • the average rate threshold of the data stream is expressed as the maximum value allowed by the average rate of the data stream within the set time.
  • the bandwidth, cache and other resources pre-allocated to the target service level are relatively In the case of insufficient, while maintaining the maximum delay associated with the target service level unchanged, increase the allocation of queue resources for the target service level, such as increasing bandwidth, buffering, and so on. It is also possible to further adjust data stream admission constraints, such as increasing the burst threshold of the data stream, the average rate threshold of the data stream, and so on. In this way, the bottleneck of the target service level can be eliminated, more data streams can be accommodated, and resource utilization can be improved.
  • the target service level of the first network device 102 if there are relatively few data flows passing through the target service level within a period of time, for example, the upper limit of the data flow admission restriction is not reached, resulting in the pre-allocation to the target service level.
  • Resources such as bandwidth and cache are relatively idle. It can be considered to reduce the resource allocation for the target service level, such as reducing bandwidth, buffering, etc., while keeping the maximum delay of the target service level unchanged. You can also further adjust the data stream admission constraints.
  • the resources released by the target service level are transferred to other service levels of the first network device, and the resource allocation of other service levels is increased, which is conducive to the accommodation of these service levels More data flow requirements will ultimately improve overall resource utilization.
  • the target service level along the route is close to or reaches the upper bound of the data stream admission constraint, which affects the further deployment of more data streams, consider ensuring that the end-to-end service can still be met Under the premise of extended demand, the data stream is switched from the target service level to the relatively more idle service level of the device or other devices to release the resources in the original location for deployment of other data streams, thereby improving the overall resource utilization.
  • the system architecture of the embodiment of the present application is shown in FIG. 3.
  • the system architecture is composed of several unit modules of the control device 101 and the first network device 102.
  • the embodiment of the present application uses the first network device 102 in FIG. 3 as an example. instruction.
  • the control device 101 includes but is not limited to a global service deployment unit 111, a global queue monitoring and analysis unit 112, a global adjustment trigger unit 113, a global configuration update calculation unit 114, a global service level switching unit 115, and a global service level maintenance unit 116 .
  • the first network device 102 includes, but is not limited to, a local queue monitoring unit 121, a local adjustment trigger unit 122, a local configuration update calculation unit 123, a local configuration update execution unit 124, and a local abnormal information reporting unit 125.
  • a local queue monitoring unit 121 For the functions of the several unit modules of the control device 101 and the first network device 102, refer to the related description of the method flow shown in FIG. 5.
  • the control device 101 and the first network device 102 are deployed in the same network device, and the network device has the functional modules or units shown in the control device 101 and the first network device 102.
  • the methods provided in the embodiments of the present application include, but are not limited to, under the global trigger QoS adjustment process, the controller determines whether to trigger the global QoS adjustment according to the admission of the global service data flow, the global queue monitoring analysis, or the local abnormal information report; if it decides When the QoS adjustment is triggered, the controller updates the service level or switches the service level of the deployed data stream.
  • the device 101 By controlling the device 101 to trigger a complete adjustment process globally, it can be ensured that when the configuration update of the service level or the service level switch of the data stream is performed, the delay degradation of the network data stream or packet loss will not be caused.
  • the method provided in the embodiments of the present application is applicable to an IP network based on statistical multiplexing of packets, which helps to improve the applicability and effectiveness of the end-to-end delay solution based on service level guarantee.
  • control device takes the control device to globally trigger the dynamic adjustment process of the service level as an example to describe the process of the method for adjusting the service level provided in the embodiment of the present application.
  • the method flow includes the following processes. Among them, the communication between the control device and the network device is realized through NETCONF.
  • the control device obtains information related to the target service level of the first network device, where the relevant information includes at least one of queue status information, remaining data flow parameters, current data flow parameters, and reporting times of abnormal information of the target service level. information.
  • the control device monitors the service level of each network device, and obtains relevant information about the target service level of the first network device, where the target service level can be understood as the monitored service level.
  • the related information includes at least one queue status information of the target service level; in another example, the related information includes remaining data flow parameters of the target service level; in another example, the related information includes The current data flow parameters of the target service level; in another example, the related information includes the number of times of reporting abnormal information of the target service level.
  • the first network device reports the queue status information of its own service level to the control device within a set period, and the control device obtains the queue status information of the target service level of the first network device.
  • the control device locally maintains the remaining data flow parameters and the current data flow parameters of the service levels of each network device, so as to obtain the remaining data flow parameters and the current data flow parameters of the first network device.
  • the current data stream parameters include, but are not limited to, the burst volume and average rate of the current data stream.
  • the remaining data stream parameters include, but are not limited to, the burst size and average rate of the remaining data stream.
  • the current data flow parameters can be reported by the network device to the control device, and the remaining data flow parameters are obtained by the control device according to the current data flow parameters.
  • the first network device reports abnormal information to the control device based on the updated values of the configuration parameters of the local resources that do not meet the target service level under the condition that the maximum delay remains unchanged.
  • the abnormal information includes queue buffer occupation and reporting that exceed the threshold. At least one of message queuing delay, message counting information, and error information.
  • the control device obtains the number of times of reporting abnormal information based on statistics of the abnormal service quality information reported by the first network device.
  • the control device Based on any information in the related information that does not meet the threshold corresponding to the any information, the control device adjusts the parameter of the target service level according to the maximum delay associated with the target service level.
  • Case 1 The relevant information of the target service level includes at least one queue status information of the target service level.
  • the control device Based on the fact that any information in the related information does not meet the threshold corresponding to any information, the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level, including:
  • Case 1A based on the fact that any queue status information in at least one queue status information is less than a lower threshold corresponding to any queue status information, the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level.
  • the control device sets one or more service levels on the ports of each controlled network device, and the control device monitors the indicators of the queue status of each service level of each network device in real time, thereby obtaining the first network device At least one queue status information of the target service level.
  • each service level is bound to one queue or multiple queues in the first network device, where the multiple queues are referred to as a set of queues.
  • the queue status information of the target service level refers to the queue status information of one or more queues in the group of queues bound to the target service level, or, The queue status information of the target service level refers to the overall queue status information of a group of queues bound to the target service level.
  • the queue status information of the target service level includes but is not limited to at least one queue status information such as buffer occupancy of the local queue, queuing delay, and packet count.
  • the buffer occupation of the queue indicates the size of the buffer occupied by the packets in the queue bound to the target service level
  • the message queuing delay indicates the queuing delay of the packets in the queue bound to the target service level, including slave messages The time interval from entering the buffer until the message is scheduled
  • the message count indicates the amount of data per unit time or the number of packets in the queue bound to the target service level.
  • the overall queue status information refers to the bound group of queues for the buffer occupation of the queue.
  • the sum of the cache occupancy of the column is to accumulate the cache occupancy of each queue in the bound group of queues, and the result obtained is used as the overall queue status information.
  • the overall queue status information refers to the maximum value of the message queuing delay of each queue in a group of bound queues, or the average value of the message queuing delay of each queue.
  • the overall queue status information refers to the sum of the packet counts of a group of bound queues, that is, accumulate the packet counts of each queue in the bound group of queues to obtain The result is used as overall queue status information.
  • the queue status information including the buffer occupancy of the local queue, packet queuing delay, and packet count as an example. If it is detected that the buffer occupancy of the local queue of the target service level is less than the lower limit of the buffer occupancy threshold, or the packet queuing delay is less than the packet
  • the lower threshold of the queuing delay, or the packet count is less than the lower threshold of the packet count triggers the control device to adjust the target service level according to the maximum delay associated with the target service level.
  • the queue status information of the target service level refers to the queue status information of each queue in the group of queues bound to the target service level
  • the lower limit of the threshold corresponding to each queue status information of the target service level can be set according to a single queue, and the lower limit of the threshold corresponding to the queue status information of each queue in a group of queues may be the same or different.
  • the queue status information for the target service level refers to the queue status information of a group of queues bound to the target service level.
  • the lower limit of the threshold corresponding to each queue status information of the target service level can be based on the number of queues in a group of queues To set up.
  • the lower limit of the threshold corresponding to each type of queue status information is not limited in the embodiment of the present application, for example, the lower limit of the buffer occupancy threshold, the lower threshold of the packet queuing delay, and the packet count.
  • the lower limit of the threshold can be set based on experience, and can also be set according to application scenarios.
  • the control device can be triggered to adjust the target service level.
  • the control device adjusts the target according to the maximum delay associated with the target service level.
  • the service level includes: based on the fact that any queue status information in the at least one queue status information is less than the lower threshold corresponding to the any queue status information within the first reference time period, the control device according to the maximum delay associated with the target service level Adjust the target service level.
  • any queue status information is less than the lower threshold corresponding to the any queue status information, then triggering the adjustment of the target service level according to the maximum delay associated with the target service level, thereby preventing queues
  • the status information is repeatedly not less than the lower limit of the threshold corresponding to the queue status information and the target service level is repeatedly adjusted, which further improves the stability and accuracy of global resource usage.
  • Case 1B based on the fact that any queue status information in at least one queue status information exceeds the upper threshold corresponding to any queue status information, the control device adjusts the target service level parameters according to the maximum delay associated with the target service level.
  • each service level of the first network device is bound to a queue or a group of queues in the first network device.
  • the queue status information of the target service level refers to the queue status information of each queue in the group of queues bound to the target service level, or the target service
  • the level of queue status information refers to the queue status information of a group of queues bound to the target service level.
  • the queue status information of the target service level includes but is not limited to at least one queue status information such as buffer occupation of the local queue, packet queuing delay, and packet count.
  • the queue status information including the buffer occupancy of the local queue, message queuing delay, and message count as an example. If it is detected that the buffer occupancy of the local queue of the target service level exceeds the upper limit of the buffer occupancy threshold, or the message queue delay exceeds the message count.
  • the upper threshold of the message queuing delay, or the message count exceeds the upper threshold of the message count triggers the control device to adjust the parameters of the target service level according to the maximum delay associated with the target service level.
  • the control device adjusts the target according to the maximum delay associated with the target service level.
  • the service level includes: based on the fact that any queue status information in the at least one queue status information exceeds the upper threshold corresponding to any queue status information within the second reference time period, the control device adjusts according to the maximum delay associated with the target service level The parameters of the target service level.
  • Case 2 The relevant information of the target service level includes the current data flow parameters of the target service level.
  • the control device Based on any information in the related information that does not meet the threshold corresponding to any information, the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level, including:
  • Case Two A The current data flow parameter based on the target service level is less than the lower threshold corresponding to the current data flow parameter, and the control device adjusts the target service level parameter according to the maximum delay associated with the target service level.
  • the current data flow parameter based on the target service level is less than the lower threshold corresponding to the current data flow parameter
  • the control device adjusts the target service level parameter according to the maximum delay associated with the target service level, including:
  • the current data flow parameters based on the target service level are less than the lower threshold corresponding to the current data flow parameters in the third reference time period, and the control device adjusts the target service level parameters according to the maximum delay associated with the target service level.
  • the current data flow parameters of the target service level are less than the lower threshold corresponding to the current data flow parameters, it is triggered to adjust the parameters of the target service level according to the maximum delay associated with the target service level, thereby It can prevent the situation that the current data flow parameters of the target service level repeatedly appear not less than the lower threshold corresponding to the current data flow parameters to repeatedly adjust the parameters of the target service level, and further improve the stability and accuracy of resource use.
  • Case Two B The current data flow parameter based on the target service level exceeds the upper threshold value corresponding to the current data flow parameter, and the control device adjusts the parameter of the target service level according to the maximum delay associated with the target service level.
  • the current data flow parameter based on the target service level exceeds the upper threshold value corresponding to the current data flow parameter
  • the control device adjusts the parameter of the target service level according to the maximum delay associated with the target service level, including:
  • the current data flow parameters based on the target service level all exceed the upper threshold value corresponding to the current data flow parameter in the fourth reference time period, and the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level.
  • the current data flow parameters of the target service level exceed the upper threshold value corresponding to the current data flow parameter, then it is triggered to adjust the parameters of the target service level according to the maximum delay associated with the target service level, thereby It can prevent that the parameters of the target service level are repeatedly adjusted due to the current data flow parameters of the target service level being less than or equal to the upper threshold corresponding to the current data flow parameters, thereby further improving the stability and accuracy of resource usage.
  • Case 3 The relevant information of the target service level includes the remaining data flow parameters of the target service level.
  • the control device Based on any information in the related information that does not meet the threshold corresponding to any information, the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level, including:
  • Case 3A The remaining data flow parameters based on the target service level exceed the upper threshold value corresponding to the remaining data flow parameters, and the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level.
  • the remaining data flow parameters based on the target service level exceed the upper threshold value corresponding to the remaining data flow parameters
  • the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level, including: The remaining data flow parameters based on the target service level all exceed the upper threshold value corresponding to the remaining data flow parameters in the fifth reference time period, and the control device adjusts the target service level according to the maximum delay associated with the target service level.
  • the target service level can be prevented from repeatedly adjusting the target service level because the remaining data flow parameters of the target service level repeatedly appear less than or equal to the upper threshold value corresponding to the remaining data flow parameters of the target service level, and the stability and accuracy of resource use can be further improved.
  • Case 3B The remaining data flow parameters based on the target service level are less than the lower threshold corresponding to the remaining data flow parameters, and the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level.
  • the remaining data flow parameters based on the target service level are less than the lower threshold corresponding to the remaining data flow parameters
  • the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level, including:
  • the remaining data flow parameters based on the target service level are all less than the lower threshold corresponding to the remaining data flow parameters in the sixth reference time period, and the control device adjusts the target service level according to the maximum delay associated with the target service level.
  • the parameters of the target service level can be prevented from repeatedly adjusting the parameters of the target service level because the remaining data flow parameters of the target service level are not less than the lower threshold corresponding to the remaining data flow parameters of the target service level, and further improve the stability of resource use And accuracy.
  • the relevant information of the target service level includes the number of times of reporting abnormal information of the target service level.
  • the control device Based on any information in the related information that does not meet the threshold corresponding to any information, the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level, including: the number of reports of abnormal information based on the target service level exceeds For the upper threshold value corresponding to the number of times of reporting abnormal information, the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level.
  • the control device adjusts the parameters of the target service level according to the maximum delay associated with the target service level, Including: the number of reports of abnormal information based on the target service level exceeds the upper threshold corresponding to the number of reports of abnormal information in the seventh time period, and the control device adjusts the target service level according to the maximum delay associated with the target service level.
  • the seventh reference time period that the number of reports of abnormal information of the target service level exceeds the upper threshold corresponding to the number of reports of abnormal information, it is triggered to adjust the parameters of the target service level according to the maximum delay associated with the target service level, thereby It can prevent that the number of reports of abnormal information of the target service level exceeds the upper limit of the threshold corresponding to the number of reports of abnormal information due to too long time, so that the parameters of the target service level are adjusted, so that the timing of adjustment does not match the actual situation, and further Improve the stability and accuracy of resource use.
  • the parameters of the target service level include queue resource parameters and data flow constraint parameters.
  • adjusting the parameters of the target service level according to the maximum delay associated with the target service level includes: determining the first update value of the data flow constraint parameter of the target service level; under the condition that the maximum delay remains unchanged, according to the target service level
  • the first update value of the data flow constraint parameter determines the first update value of the queue resource parameter of the target service level; then, the previous data flow constraint parameter is adjusted according to the first update value of the data flow constraint parameter, and the data flow constraint parameter is adjusted according to the first update value of the queue resource parameter.
  • the first update value of the queue resource parameter includes the first update value of the queue bandwidth and the first update value of the queue buffer
  • the first update value of the data flow constraint parameter includes the maximum value of the burst volume threshold of the data flow. The first update value and the first update value of the average rate threshold of the data stream.
  • the first update value of the queue resource parameter is determined according to the first update value of the data flow constraint parameter, including: under the condition that the maximum delay is unchanged, determining the port corresponding to the target service level Bandwidth, the first maximum amount of data that can be forwarded by the target service level after the parameters of the target service level are adjusted, the maximum amount of data that can be forwarded by the multiple service levels of the first network device before the parameters of the target service level are adjusted, and multiple service levels
  • the maximum message length of the corresponding queue according to the bandwidth of the port corresponding to the target service level, the first maximum amount of data that can be forwarded after the parameter adjustment of the target service level at the target service level, and the parameter adjustment of multiple service levels at the target service level
  • the maximum amount of data that can be forwarded before determines the first update value of the queue bandwidth; the maximum amount of data that can be forwarded before the parameter adjustment of the target service level according to multiple service levels and the maximum message length of the queue corresponding to multiple service levels,
  • the first maximum amount of data that can be forwarded by the target service level after the parameters of the target service level are adjusted, and the maximum amount of data that can be forwarded by multiple service levels before the parameters of the target service level are adjusted
  • Data volume determine the first update value of the queue bandwidth, including: according to the bandwidth of the port corresponding to the target service level, the first maximum data volume that can be forwarded after the target service level is adjusted in the target service level parameters, and multiple service levels in the target
  • the maximum amount of data that can be forwarded before the parameters of the service level are adjusted, and the first update value of the queue bandwidth is determined according to the following formula
  • C is the bandwidth of the port corresponding to the target service level
  • Is the first maximum data volume that can be forwarded by the target service level after the parameters of the target service level are adjusted
  • Q j is the maximum data volume that can be forwarded by the j-th service level
  • It is the maximum amount of data that the non-delay guarantee queue can forward before the parameters of the target service level are adjusted
  • n is a positive integer greater than 1.
  • the control device configures the maximum amount of data that can be forwarded by the target service level before adjusting the parameters of the target service level this time.
  • the Adjustments the That is, the maximum amount of data that can be forwarded by the target service level before the parameters of the target service level are adjusted.
  • the control device configures the maximum amount of data that can be forwarded by the non-delay guarantee queue before adjusting the parameters of the target service level this time, then for this adjustment, the That is, the maximum amount of data that can be forwarded by the non-delay guarantee queue before the parameters of the target service level are adjusted.
  • the control equipment will adjust the and Has been stored, then for this adjustment, get the stored As the maximum amount of data that can be forwarded by the target service level before the parameters of the target service level are adjusted, obtain the stored As the maximum amount of data that the non-delay guarantee queue can forward before the parameters of the target service level are adjusted.
  • the control device adjusts the parameter of the jth service level, and the control device may also store the adjusted value. Then, for the parameter of this adjustment of the target service level, the stored Q j can be directly obtained as the maximum amount of data that can be forwarded by the jth service level used in the parameter adjustment of the current target service level.
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level, the maximum message length of the queues corresponding to the multiple service levels, the first update value of the burst threshold of the data flow, and The first update value of the average rate threshold of the data flow, which determines the first update value of the queue buffer, includes:
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level, the maximum packet length of the queues corresponding to multiple service levels, the first update value of the burst threshold of the data flow, and the average of the data flow is determined according to the following formula
  • L max L is the maximum packet length in the low priority queue
  • C is the bandwidth of the port corresponding to the target service level
  • Q j is the maximum data volume that can be forwarded by the j-th service level
  • L n+1 is the maximum packet in the non-delay guarantee queue Length
  • L j is the maximum packet length in the j-th service level queue
  • ⁇ b i1 is the decrease value of the burst threshold of the data stream of the target service level
  • the maximum packet length L i is the target service-level queue
  • L n+1 , L j , Li , L max, L and C may be pre-stored on the control device. For example, after the control device allocates bandwidth to the port corresponding to the target service level, L n +1 , L j , Li , L max, L and C, etc. are stored. Then the control device directly obtains the stored L n+1 , L j , Li , L max, L and C.
  • the configuration parameters of the target service level include queue resource parameters and data flow constraint parameters; adjust the parameters of the target service level according to the maximum delay associated with the target service level, Including but not limited to the adjustment method in the first case where the second update value of the queue resource parameter is less than or equal to the resource threshold, and the adjustment method in the second case where the second update value of the queue resource parameter is greater than the resource threshold.
  • the adjustment methods for the two cases are as follows.
  • the adjustment method in the first case determine the second update value of the data flow constraint parameter; under the condition that the maximum delay remains unchanged, determine the second update value of the queue resource parameter according to the second update value of the data flow constraint parameter; Based on the second update value of the queue resource parameter being less than or equal to the resource threshold, the admission restriction parameter of the previous data flow is adjusted according to the second update value of the data flow restriction parameter, and the previous queue resource parameter is adjusted according to the second update value of the queue resource parameter.
  • the second update value of the queue resource parameter includes the second update value of the queue bandwidth and the second update value of the queue buffer
  • the second update value of the data flow constraint parameter includes the maximum value of the burst volume threshold of the data flow.
  • the second update value and the second update value of the average rate threshold of the data stream includes the second update value and the second update value of the average rate threshold of the data stream.
  • the second update value of the queue resource parameter is determined according to the second update value of the data flow constraint parameter, including: under the condition that the maximum delay is unchanged, determining the port corresponding to the target service level Bandwidth, the second maximum amount of data that can be forwarded by the target service level after the parameters of the target service level are adjusted, the maximum amount of data that can be forwarded by the multiple service levels of the first network device before the parameters of the target service level are adjusted, and multiple service levels
  • the maximum message length of the corresponding queue according to the bandwidth of the port corresponding to the target service level, the second maximum amount of data that can be forwarded after the parameter adjustment of the target service level at the target service level, and the parameter adjustment of multiple service levels at the target service level
  • the maximum amount of data that can be forwarded before determines the second update value of the queue bandwidth; the maximum amount of data that can be forwarded before the parameter adjustment of the target service level according to multiple service levels and the maximum packet length of the queue corresponding to multiple service levels,
  • the second maximum amount of data that can be forwarded by the target service level after the parameters of the target service level are adjusted, and the maximum amount of data that can be forwarded by multiple service levels before the parameters of the target service level are adjusted
  • the data volume determines the second update value of the queue bandwidth, including: according to the bandwidth of the port corresponding to the target service level, the second maximum data volume that can be forwarded after the target service level is adjusted in the target service level parameters, and the target service level of each service level
  • the maximum amount of data that can be forwarded before the level parameter is adjusted, and the second update value of the queue bandwidth is determined according to the following formula
  • C is the bandwidth of the port corresponding to the target service level
  • Q j is the maximum data volume that can be forwarded by the j-th service level
  • n is a positive integer greater than 1.
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level, the maximum packet length of the queues corresponding to the multiple service levels, the second update value of the burst threshold of the data flow, and The second update value of the average rate threshold of the data flow determines the second update value of the queue cache, including: the maximum amount of data that can be forwarded before the parameters of the target service level are adjusted according to multiple service levels and the queues corresponding to the multiple service levels.
  • the second update value of the maximum packet length of the data stream, the second update value of the burst threshold of the data stream, and the second update value of the average rate threshold of the data stream determine the second update value of the queue buffer according to the following formula
  • C is the bandwidth of the port corresponding to the target service level
  • Q j is the maximum amount of data that can be forwarded by the j-th service level
  • L max L is the maximum packet length in the low priority queue
  • L n+1 is the maximum message length in the non-delay guarantee queue
  • L j is the maximum message length in the j-th service level queue
  • ⁇ b i2 is the increase value of the burst threshold of the data stream
  • the maximum packet length L i is the target service-level queue
  • n is an integer greater
  • the adjustment method 1 The configuration parameters of the target service level include queue resource parameters and data flow constraint parameters; the parameters of the target service level are adjusted according to the maximum delay associated with the target service level, including: data to determine the target service level The second update value of the flow constraint parameter; under the condition that the maximum delay is unchanged, the second update value of the queue resource parameter of the target service level is determined according to the second update value of the data flow constraint parameter of the target service level; based on the queue resource The second update value of the parameter is greater than the resource threshold, and the target data stream corresponding to the target service level is switched to another service level, and the other service levels include other service levels of the first network device or service levels of other devices.
  • the resource threshold includes but is not limited to the size of the resource that can be preempted by the non-delay guarantee queue, and the resource includes but is not limited to bandwidth and cache.
  • the second update value of the data flow restriction parameter includes the second update value of the burst volume threshold of the data flow and the second update value of the average rate threshold of the data flow; determined according to the second update value of the data flow restriction parameter
  • the third update value of the queue resource parameter includes: determining the bandwidth of the port corresponding to the target service level, the third maximum amount of data that the target service level can forward after the parameters of the target service level are adjusted, and the number of first network devices The maximum amount of data that the service level can forward before the parameter adjustment of the target service level and the maximum message length of the queues corresponding to multiple service levels; according to the bandwidth of the port corresponding to the target service level, the target service level is within the target service level
  • the third maximum amount of data that can be forwarded after parameter adjustment and the maximum amount of data that can be forwarded by multiple service levels before the parameters of the target service level are adjusted to determine the third update value of the queue bandwidth; The maximum amount of data that can be forwarded before parameter adjustment and the maximum message length of the queues corresponding to multiple service levels
  • Determine the updated value of the parameters of the service level of the preemptible resource including: according to the bandwidth of the port corresponding to the target service level, the maximum amount of data that can be forwarded after the adjustment of the service level of the preemptable resource, and the parameters of multiple service levels at the target service level
  • the maximum amount of data that can be forwarded before adjustment determine the updated value of the queue bandwidth of the service level that can grab resources; the maximum amount of data that can be forwarded before the parameter adjustment of the target service level according to multiple service levels, and the queues corresponding to multiple service levels
  • the updated value of determines the updated value of the queue cache of the service level that can preempt resources.
  • the third largest amount of data that can be forwarded by the target service level after the parameters of the target service level are adjusted, and multiple service levels can be forwarded before the parameters of the target service level are adjusted Determine the third update value of the queue bandwidth, including: according to the bandwidth of the port corresponding to the target service level, the third maximum amount of data that can be forwarded after the target service level is adjusted after the parameters of the target service level are adjusted, and multiple The maximum amount of data that can be forwarded by the service level before the parameters of the target service level are adjusted, and the third update value of the queue bandwidth is determined according to the following formula
  • the maximum amount of data that can be forwarded after the adjustment of the service level that can preempt resources determines the service that can preempt resources Level of queue bandwidth
  • C is the bandwidth of the port corresponding to the target service level
  • Q j is the maximum amount of data that can be forwarded by the jth service level
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level, the maximum packet length of the queues corresponding to the multiple service levels, the second update value of the burst threshold of the data flow, and The second update value of the average rate threshold of the data flow, which determines the third update value of the queue buffer, includes:
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level the maximum packet length of the queues corresponding to multiple service levels, the maximum amount of data that can be forwarded after the adjustment of the service level that can preempt resources, and the maximum report
  • the second update value of the text length, the burst threshold of the data stream, and the second update value of the average rate threshold of the data stream determine the third update value of the queue buffer according to the following formula
  • the maximum amount of data that can be forwarded before the parameter adjustment of the target service level the maximum packet length of the queues corresponding to multiple service levels, the maximum amount of data that can be forwarded after the adjustment of the service level that can preempt resources, and the maximum report
  • the length of the text, the update value of the burst threshold value of the data flow of the service level that can preempt resources, and the update value of the average rate threshold of the data flow determine the queue cache of the service level of the resource that can be preempted
  • C is the bandwidth of the port corresponding to the target service level
  • Q j is the maximum amount of data that can be forwarded for the jth service level
  • L max L is the maximum packet length in the low priority queue
  • L n+1 is the maximum message length in the non-delay guarantee queue
  • L j is the maximum message length in the j-th service level queue
  • ⁇ b i2 is the increase value of the burst threshold of the data stream
  • the maximum packet length L i is the target service-level queue
  • L k to be preempted resources
  • the first adjustment method and the second adjustment method in the second case described above are both the processing method when the second update value of the queue resource parameter is greater than the resource threshold.
  • the queue resource parameter In the case where the second update value of is greater than the resource threshold, whether the first adjustment method or the second adjustment method is adopted, which is not limited in the embodiment of the present application. That is to say, when the second update value of the queue resource parameter is greater than the resource threshold, the adjustment method 1 can be directly used to switch the target data flow corresponding to the target service level to other service levels, or the adjustment method 2 can be used.
  • the second update value of the data flow restriction parameter adjusts the data flow restriction parameter
  • the queue resource parameter is adjusted according to the third update value of the queue resource parameter
  • the service level parameter of the preemptible resource is adjusted according to the updated value of the service level parameter of the preemptible resource .
  • the adjustment method 2 can be continued.
  • the third adjustment method may be used.
  • control device determines the updated value of the parameter of the target service level
  • the updated value of the parameter of the service level adjusts the parameter of the target service level.
  • the first network device receives the updated value of the parameter of the target service level sent by the control device; and adjusts the parameter of the target service level according to the updated value of the parameter of the target service level.
  • the update value also includes: the control device sends the update value of the parameter of the adjustment of the service level other than the target service level to the first network device, and instructs the first network device to update the parameter of the service level other than the target service level.
  • Value adjustment corresponds to other service levels.
  • the first network device receives updated values of parameters of other service levels other than the target service level sent by the control device; and adjusts corresponding other service levels according to the updated values of the parameters of other service levels.
  • the updated value of the queue buffer in the embodiment of the present application refers to the minimum required value of the queue buffer.
  • the value greater than Other values of the update value of the queue buffer are not limited in the embodiment of the present application.
  • the method provided by the embodiment of the application dynamically updates the queue resource allocation and data flow admission constraints based on changes in service deployment conditions (for example, current data flow parameters and remaining data flow parameters), queue status information and the number of times of reporting abnormal information, etc.
  • the method for adjusting the service level provided by the embodiment of the present application will be described through the interaction process between the first network device and the controller. As shown in Figure 5, the method includes the following 501-511 processes.
  • the controller global service deployment unit 111 receives the service requirements, calculates the path and the service level that needs to be entered along the path for the service, and deploys it. At the same time, maintain the current data flow parameters and the remaining allowable access data flow size for all service levels of the network, including the current data flow parameters, according to the corresponding parameters of the new service contract promised data flow (data flow burst threshold, data flow The average rate threshold of the flow, etc.) is increased, and the parameters of the current remaining allowable admission flow are deducted according to the corresponding parameters of the new service contract commitment data flow.
  • data flow burst threshold data flow The average rate threshold of the flow, etc.
  • the local queue monitoring unit 112 of each network device periodically reports various local queue status information to the global queue monitoring and analyzing unit 121, and the global queue monitoring and analyzing unit 121 maintains the queue status information of all service levels globally.
  • the local queue monitoring unit 111 of the first network device periodically reports queue status information such as local queue buffer occupancy, message queuing delay, or message count to the controller global queue monitoring analysis unit 121.
  • the network device local abnormal information reporting unit 125 reports the local QoS adjustment abnormal information to the controller in real time.
  • the global adjustment trigger unit 113 triggers the configuration update or trigger of the global target service level according to the current data flow admission status of the entire network service level, global queue monitoring information, and local QoS abnormal information reporting status The level of service through which the related deployed data stream is switched.
  • triggering the configuration update of the global target service level, or triggering the service level of the related deployed data stream switching includes but not limited to the following seven situations.
  • the first scenario According to the queue monitoring information, if any queue status information such as the queue buffer occupancy of the target service level, the message queuing delay, and the message count is less than the corresponding lower threshold within the first reference time period, the global The adjustment triggering unit 113 judges that the actual arrival size of the data stream in the target service level is less than the maximum data stream size contracted by the target service level for a long time, and then triggers the queue resources of the target service level, data stream admission and other parameters to be configured in a decreasing direction renew.
  • queue status information such as the queue buffer occupancy of the target service level, the message queuing delay, and the message count is less than the corresponding lower threshold within the first reference time period
  • the second situation If the current data flow parameters of the target service level (the burst volume threshold of the data flow and the average rate threshold of the data flow) are less than the lower threshold corresponding to the current data flow parameter in the third reference time period, the global The adjustment triggering unit 113 determines that the data flow deployment in the target service level has been in an insufficient state for a long time, and then triggers the configuration update of parameters such as queue resources and data flow admission of the target service level in a decreasing direction.
  • the third situation if the remaining data flow parameters of the target service level (the maximum remaining allowable burst and the maximum remaining average rate) both exceed the upper threshold corresponding to the remaining data flow parameters within the fifth reference time period, the global adjustment trigger unit 113 It is judged that the data flow in the target service level is close to the lower limit of deployment, which then triggers the configuration update of the queue resources and data flow admission of the target service level in the decreasing direction.
  • the fourth situation according to the queue monitoring information, if any queue status information such as the queue buffer occupation of the target service level, the message queuing delay, and the message count is greater than the corresponding upper threshold within the second reference time period, the global The adjustment trigger unit 113 determines that the data flow actually arriving in the target service level exceeds the data flow admission restriction of the target service level, and then triggers the configuration update of the queue resources and data flow admission of the target service level in the increasing direction, or Trigger the deployed data stream passing the target service level to switch to the other service levels of the first network device or the service levels of other network devices.
  • the global The adjustment trigger unit 113 determines that the data flow actually arriving in the target service level exceeds the data flow admission restriction of the target service level, and then triggers the configuration update of the queue resources and data flow admission of the target service level in the increasing direction, or Trigger the deployed data stream passing the target service level to switch to the other service levels of the first network device or the service levels of other network devices.
  • the fifth situation If the current data flow parameters of the target service level (the burst threshold of the data flow or the average rate of the data flow) exceed the upper threshold corresponding to the current data flow parameter in the fourth reference time period, the global The adjustment triggering unit 113 determines that the data flow deployment in the target service level is close to the deployment upper limit, and then triggers the configuration update of the queue resources of the target service level, data flow admission and other parameters in the increasing direction, or triggers the deployed data that has passed the target service level The flow is switched to the other service level of the first network device or the service level of the other network device.
  • the sixth situation if the remaining data flow parameters (maximum remaining allowable burst or maximum remaining average rate) of the target service level are less than the lower threshold corresponding to the remaining data flow parameters in the sixth reference time period, the global adjustment trigger unit 113 Determine that the data flow in the target service level is close to the upper limit of deployment, and then trigger the configuration update of the queue resources and data flow admission of the target service level in the increasing direction, or trigger the deployed data flow that has passed the target service level to switch to the first The other service levels of a network device or the service levels of other network devices.
  • the seventh scenario If the number of times of receiving the target service level data flow over-admission restriction ERROR signal report exceeds the threshold upper limit corresponding to the number of data flow over-admission restriction ERROR signal report times of the target service level, the global adjustment trigger unit 113 determines the target The data flow actually reached by the service level exceeds the data flow admission constraint of the target service level, and cannot be processed locally by adjusting the target service level by itself, which triggers the increase of the queue resources and data flow admission parameters of the target service level Perform a configuration update, or trigger the deployed data stream that has passed the target service level to switch to the other service level of the first network device or the service level of the other network device.
  • the network-wide adjustment calculation unit 114 calculates the updated configuration of the network-wide related service levels, including new queue resource parameters (queue bandwidth, queue buffer, etc.) and data flow constraint parameters (data flow burst threshold, data The average rate threshold of the flow, etc.); or calculate the new service level that needs to be passed by each hop along the route for the related deployed data flow of the entire network, in order to meet the end-to-end delay requirements of the business, and the data flow admission of each service level Under the premise of restriction, you can either choose to switch the original service level to another service level of the same port, or switch it to another service level of the same device with a different port or even a different device.
  • new queue resource parameters queue bandwidth, queue buffer, etc.
  • data flow constraint parameters data flow burst threshold, data The average rate threshold of the flow, etc.
  • the global adjustment calculation unit 114 calculates the update configuration of the related service level of the entire network including but not limited to the following two methods.
  • the global adjustment calculation unit 114 calculates the data flow admission that needs to be reduced for the relevant service level (including the burst threshold of the data flow that needs to be reduced, and the data flow The average rate threshold of the flow, etc.), while keeping the associated maximum delay unchanged, calculate the queue resources that need to be reduced corresponding to the target service level (including the queue bandwidth that needs to be reduced, queue buffers, etc.).
  • the global adjustment calculation unit 114 calculates the updated configuration of the service level based on the network calculation theory, the principle is as shown in the schematic diagram of the egress port queue as shown in FIG.
  • Each is bound to a queue.
  • a virtual queue is used to refer to these no-delay guarantee demand queues, which is called a non-delay guarantee queue. All queues corresponding to service levels and non-delay guarantee queues are scheduled in a round-robin manner.
  • the queues corresponding to the service levels and the non-delay guarantee queues are high-priority queues. Under each scheduling opportunity, the high-priority queue can be prioritized for scheduling.
  • Target service level for the i-th service level for example, round robin scheduling when the high-priority queue inside of the turn of the i-th service level, the maximum amount of data scheduled to be forwarded Q i, when the non-turn When the guarantee queue is delayed, the maximum amount of data that can be forwarded in one scheduling is Q n+1 .
  • Q n+1 0 can be directly set.
  • the independent variable is t, in the interval t 2 -t 1 between t 1 and t 2 (t 2 ⁇ t 1 ⁇ 0) at any time ,
  • the amount of data successfully forwarded by this service level is not less than ⁇ i (t 2 -t 1 ).
  • D i the maximum delay requirement of the i-th service class
  • Li the maximum message length in the queue corresponding to the i-th service class
  • the maximum message length in the non-delay guarantee queue is L n+1
  • the maximum message length in the low priority queue is L max,L .
  • the maximum amount of data that can be forwarded each time it is the i-th service level before the configuration update is The maximum amount of data that can be forwarded by the non-delay guarantee queue is In particular, if the non-delay guarantee queue does not exist before the configuration update, let Before the i-th service level configuration is updated, the egress bandwidth of the i-th service level is Correspondingly, before the i-th service level configuration is updated, the burst threshold of the data flow of the i-th service level is The average rate is As shown in (1) in Figure 7, Represents the arrival curve of the i-th service level before the update, at this time the service curve of the i-th service level Expressed as follows:
  • T 0 is the time offset term
  • the maximum delay D i is the transmission data stream delay of a single network device including the ingress port queuing delay, the egress port queuing delay, the processing delay, and the sending delay as an example.
  • T 0 is defined as the sum of the ingress port queuing delay, processing delay, and sending delay of the network device; taking the maximum delay Di as the egress port queuing delay as an example, T 0 is defined as 0.
  • the maximum delay value of the i-th service class is D i expressed as follows:
  • the i-th service level is the minimum queue buffer size required to meet the maximum delay value D i Expressed as follows:
  • the global adjustment calculation unit 114 calculates that the burst threshold of the data stream that needs to be reduced for the i-th service level is ⁇ b i , and the first updated value of the updated burst threshold of the data stream is The reduction value of the average rate threshold of the data stream that needs to be reduced is ⁇ r i , and the first update value of the average rate threshold of the updated data stream is After the i-th service level configuration is updated, as shown in (1) in Figure 7, Represents the service curve after the i-th service level is updated, The abscissa intercept is Denotes the i th updated service level reaches the curve, this time to maintain the service level of the i-th value of the maximum delay D i constant, can be re-expressed as follows:
  • the first maximum data volume that can be forwarded when it is the newly calculated turn to schedule the i-th service level that is, the first maximum data volume that can be forwarded by the target service level after the parameters of the target service level are adjusted; It is the newly calculated maximum amount of data that can be forwarded when it is the turn of the non-delay guarantee queue, that is, the maximum amount of data that the non-delay guarantee queue can forward after the parameters of the target service level are adjusted.
  • formula (1-5) the first maximum amount of data that can be forwarded when the i-th service level is scheduled after the configuration update can be calculated as follows:
  • the minimum required queue buffer size of the i-th service level is the first updated value of the queue buffer
  • L max L is the maximum packet length in the low priority queue
  • C is the bandwidth of the port corresponding to the target service level
  • Q j is the maximum data volume that can be forwarded by the j-th service level
  • L n+1 is the maximum packet in the non-delay guarantee queue Length
  • L j is the maximum packet length in the j-th service level queue
  • ⁇ b i1 is the decrease value of the burst threshold of the data stream of the target service level
  • the maximum packet length L i is the target service-level queue
  • the local configuration update execution unit 124 preempts The corresponding resources of the non-delay guarantee queue are transferred to the target service level, and the parameters of the target service level are updated according to the new queue bandwidth and cache requirements.
  • the global adjustment calculation unit 114 calculates the additional data flow access required for the relevant service level (including the data flow access that needs to be increased).
  • the burst threshold, the average rate threshold of the data flow, etc. while keeping the maximum delay associated with the target service level unchanged, the calculation of the target service level corresponding to the need to increase the queue resources (including the need to increase the reserved bandwidth of the queue , Caching, etc.).
  • the calculation result needs to be guaranteed: the increased queue resources of the target service level does not exceed the non-delay guarantee queue resources (bandwidth, buffer) that can be preempted under the same port and other service levels under the same port without affecting its own maximum delay requirements The sum of the releasable queue resources.
  • the global adjustment calculation unit 114 calculates the updated configuration of the service level based on the network calculation theory.
  • the principle is shown in the port diagram of Figure 6.
  • the egress bandwidth of the i-th service level is Correspondingly, before the i-th service level configuration is updated, the burst threshold of the data flow of the i-th service level is The average rate is As shown in (2) in Figure 7, Represents the arrival curve of the i-th service level before the update, at this time the service curve of the i-th service level Expressed as follows:
  • T 0 is the time offset term
  • the maximum delay D i is the transmission data stream delay of a single network device including the ingress port queuing delay, the egress port queuing delay, the processing delay, and the sending delay as an example.
  • T 0 is defined as the sum of the ingress port queuing delay, processing delay, and sending delay of the network device; taking the maximum delay Di as the egress port queuing delay as an example, T 0 is defined as 0.
  • the maximum delay value of the i-th service class is D i expressed as follows:
  • the i-th service level is the minimum queue buffer size required to meet the maximum delay value D i Expressed as follows:
  • the global adjustment calculation unit 114 calculates that the burst threshold of the data stream that needs to be increased for the i-th service level is ⁇ b i2 , and the second updated value of the updated burst threshold of the data stream is The average rate threshold of the data stream that needs to be increased is ⁇ r i2 , and the second updated value of the average rate threshold of the updated data stream is
  • the resources (bandwidth, buffer) of the non-delay guarantee queue that can be preempted under the same port are preempted first. If the resources of the non-delay guarantee queue that can be preemptible under the same port are not Meet the requirements, and further seize the queue resources that can be released by other service levels on the same port without affecting its own maximum delay requirements.
  • the target service level directly preempts the least amount of non-timeliness that can meet the adjustment requirements Delay guarantee queue resources; 1b) If the preemptible resources of the non-delay guarantee queue are insufficient to meet the target service level adjustment requirements, the target service level first preempts all preemptible resources in the non-delay guarantee queue, and then successively preempts other service levels on the same port. The queue resources that can be released under the premise of not affecting their own maximum delay requirements, until the adjustment requirements of the target service level are met. 2) If the non-delay guarantee queue does not exist on the same port, it directly preempts the queue resources that can be released by other service levels on the same port without affecting its own maximum delay requirements until the adjustment requirements of the target service level are met.
  • Case 1 There is a non-delay guarantee queue.
  • the maximum delay value D i of the i-th service level remains unchanged, it is assumed that in order to meet the updated traffic admission (the second updated value of the burst threshold of the data flow The second updated value of the average rate threshold of the data stream ), the maximum amount of data that can be forwarded when it is the turn to schedule the i-th service level needs to be updated to Correspondingly, the maximum amount of data that can be forwarded when it is the turn to schedule the non-delay guarantee queue needs to be updated to For example, after the i-th service level configuration is updated, as shown in (2) in Figure 7, Represents the service curve after the i-th service level is updated, The abscissa intercept is Represents the arrival curve after the i-th service level is updated, and D i can be re-expressed as follows:
  • formula (2-3) calculate the minimum required queue buffer size for the i-th service level assuming that the adjustment requirements are met Need to be updated to:
  • Case 1 A Define the current maximum preemptible bandwidth resource of the non-delay guarantee queue as ⁇ C n+1 , and the maximum preemptible buffer resource as ⁇ BF n+1 . If the current preemptible resources of the non-delay guarantee queue meet the adjustment requirements of the target service level, the following two conditions are met:
  • the second updated value of the queue bandwidth of the target service level can be determined At the same time determine the second update value of the queue cache
  • the second updated value of the queue bandwidth And the second updated value of the queue cache For the determination process, please refer to the adjustment method in the first case in 402, the second update value of the queue bandwidth
  • the determination process can refer to the above The process of determining.
  • the determination process can refer to the above The process of determining.
  • Case 1B If the current preemptible resources of the non-delay guarantee queue do not meet the adjustment requirements of the target service level, adjust the target service level according to the maximum delay associated with the target service level, including: determining the second update value of the data flow constraint parameter; When the maximum delay is unchanged, the second update value of the queue resource parameter is determined according to the second update value of the data flow constraint parameter; based on the second update value of the queue resource parameter being greater than the resource threshold, the target data corresponding to the target service level is determined The flow is switched to other service levels, and the other service levels include other service levels of the first network device or service levels of other devices.
  • the target service level first preempts all preemptible resources of the non-delay guarantee queue, and then sequentially preempts the queue resources that can be released by other service levels on the same port without affecting its own maximum delay requirements, until the adjustment of the target service level is satisfied Require.
  • the current preemptible buffer resources of the non-delay guarantee queue are large enough to always meet the adjustment requirements of the target service level.
  • the following mainly describes the adjustment process of bandwidth resources. Considering that the current preemptible bandwidth resources of the non-delay guarantee queue are insufficient and cannot meet the adjustment requirements of the target service level, that is, there are:
  • the bandwidth resource of the target service level needs to be adjusted to
  • the maximum amount of data that can be forwarded when it is the turn to dispatch the target service level at this time needs to be adjusted to:
  • preempting the bandwidth resources of the non-delay guarantee queue cannot meet the adjustment requirements of the target service level, the adjustment parameters of the target service level at this time and It is only a temporary state. It is necessary to further consider preempting the queue resources that can be released by other service levels on the same port without affecting its own maximum delay requirements.
  • preempting the non-delay guarantee queue or preempting the corresponding queue resources of other queues under the same port is regarded as an iterative process. Therefore, preempting the corresponding queue resources of the non-delay guarantee queue is regarded as the first iteration process, which is a unified expression.
  • the current service level parameters Q j and the non-delay guarantee queue parameters are maintained constant.
  • the total traffic admission parameter before the update of the kth service level as and They are the burst threshold of the data flow before the update and the average rate threshold of the data flow.
  • the current allowable access flow parameter of the kth service level as and They are the current allowable admission burst threshold and the current allowable admission average rate threshold. According to the definition, there are It should be pointed out that the selection method of other service levels is not limited in the embodiment of this application, and may be based on experience or certain rules, such as preferential selection of relatively more idle service levels.
  • the maximum amount of data that can be forwarded when it is the turn to schedule the i-th service level that needs to be updated is defined as
  • the maximum amount of data that can be forwarded when it is the turn to schedule the k-th service level to be updated is defined as Analogous to formula (2-5), in order to ensure that the maximum delay of other service levels is not affected when the i-th service level and the k-th service level are adjusted, the following constraints need to be met:
  • the maximum delay of the current allowable access traffic for the k-th service level still meets the requirements, that is, the following conditions are met:
  • the maximum amount of data that can be forwarded when it is the turn to dispatch the target service level can be finally determined Finally determine the third updated value of the queue bandwidth of the target service level for:
  • the third update value of the queue cache of the target service level is finally determined for:
  • the maximum delay of the current allowable access traffic for the k-th service level cannot meet the requirements, that is, the following conditions are met:
  • the maximum forwarded data volume of the scheduled target service level is updated to Under the second iteration process, the burst threshold of the data flow of the target service level admission traffic Need to meet:
  • the order of the j Kth service level is adjusted iteratively.
  • the maximum amount of data that can be forwarded when it is the turn of the scheduling target service level is The maximum amount of data that can be forwarded when it is the turn of the non-delay guarantee queue is The maximum amount of data that can be forwarded when it is the turn of the other j kth service level is Where 1 ⁇ k ⁇ K
  • the third updated value of the queue bandwidth of the target service level is finally determined for:
  • the third update value of the queue cache of the target service level is finally determined for:
  • Is the updated value of the average rate threshold of the data flow of the j kth service class Is the updated value of the burst threshold value of the data flow of the j kth service class, in Is the quantization symbol, Is the average rate threshold of the j kth service class currently allowed admission, J k are the classes of service currently allowed access burst threshold.
  • the maximum amount of data that can be forwarded when it is the turn to schedule the non-delay guarantee queue The basic calculation process is the same as situation 1, but the first iteration process of preempting non-delay guarantee queue resources is omitted, and the second iteration process of preempting resources of other service levels under the same port is directly started. Go into details.
  • the deployed data stream passes through some data streams that are close to the deployment limit or the target service level of the data stream being over-delivered, and the global adjustment calculation unit 114 calculates a new service level to be passed by each hop along the route for these data streams.
  • the deployment according to the result still meets the end-to-end delay requirements of the business and the data flow admission constraints of each service level.
  • priority is given to switching to other relatively idle service levels of the same port on the local device. If there is no feasible calculation result, further consider switching to other ports of the local device or other service levels where other network devices are relatively idle.
  • the global adjustment calculation unit 114 causes the data stream service level switching to be performed before the service level update configuration. If it is calculated that the additional queue resources (including queue bandwidth, cache, etc.) that need to be added to the target service level exceed the non-delay guarantee queue resources (bandwidth, cache) that can be preempted under the same port, and other service levels under the same port can release queue resources Then, the service level that needs to be switched is further calculated for the related deployed data streams passing through other service levels, so that other service levels can release queue resources to increase, so as to meet the additional queue resource requirements of related service levels. It is stipulated that before performing the above-mentioned service level configuration update, the target service level switching of the above-mentioned deployed data stream is performed in advance.
  • the global adjustment calculation unit 114 causes the target service level update configuration to be executed before the data stream service level switching. If the path switching result of the relevant deployed data stream is calculated, but the remaining data stream access of the service level to be switched to is insufficient, the configuration is further calculated and updated for the service level, so that the target service level can meet the switching requirements of the relevant data stream . It is stipulated that the configuration update of the target service level is performed in advance before the service level switching of the deployed data stream is performed.
  • the controller delivers the target service level update configuration result calculated by the entire network adjustment calculation unit 114 to the local configuration update execution unit 124 of the relevant device.
  • the controller also sends the corresponding updated values of the parameters of other service levels to the local configuration update execution unit 124 of the relevant device.
  • the service level update configuration information includes new queue resource parameters (queue reserved bandwidth, queue buffer, etc.), data flow restriction parameters (data flow burst threshold, data flow average rate threshold, etc.).
  • the local configuration update execution unit 124 implements a corresponding configuration update to the target service level.
  • the local configuration update execution unit 124 implements the configuration update of the target service level based on the calculation result of the first method in 505 above, and prioritizes the bandwidth of the non-delay guarantee queue that can be preempted under the same port, The cache resources are transferred to the target service level. If the additional queue resource requirements of the target service level are still not met, continue to consider transferring the queue resources that can be reduced from other relatively idle service levels on the same port to the target service level without affecting the maximum delay requirements of the port, until it is achieved. Specify the required configuration update.
  • the local configuration update execution unit 124 implements the configuration update of the target service level based on the calculation result of the second method in 505 above, and transfers the queue resources released by the target service level to those that can be preempted under the same port. Non-delay guarantees the bandwidth and buffer resources of the queue.
  • the global service level maintenance unit 116 synchronizes the updated values of the parameters of the target service level.
  • the subsequent controller performs service deployment based on the updated values of the parameters of the target service level.
  • the global service level switching unit 115 switches the service levels that the related data flow needs to pass through according to the service level switching result calculated by the network-wide adjustment calculation unit 114, and subsequent packets of related data flows are performed according to the new service level transmission.
  • the global service deployment unit 111 modifies the contractual commitment data flow parameters of related services as needed according to the actual arrival situation, including modifying the burst threshold of the data flow, the average rate threshold of the data flow, etc., combined with For the update configuration of the relevant service level or the service level switching of the relevant business, the current allowable data stream size and the remaining allowable data stream size of each service level are modified as needed.
  • the global service deployment unit 111 upwardly corrects the parameters of the data stream that exceeds the contract commitment within the target service level according to the actual arrival situation, including upwardly correcting the burst threshold of the data stream and the average rate threshold of the data stream.
  • the long-term parameters of the data stream that are less than the contract commitment within the target service level, such as the target reference time period are revised downwards according to the actual arrival situation, including the downward revision of the burst threshold of the data stream and the average rate of the data stream. Wait.
  • An embodiment of the present application provides an apparatus for adjusting a service level, which implements the method for adjusting the service level shown in FIGS. 4 and 5 through the module shown in FIG. 8.
  • the device includes:
  • the obtaining module 801 is configured to obtain relevant information of the target service level of the first network device, and the relevant information includes at least one of queue status information, remaining data flow parameters, current data flow parameters, and reporting times of abnormal information of the target service level.
  • the functions performed by the acquisition module 801 may refer to the related content described in 401 in FIG. 4, or refer to the related content described in 501-503 shown in FIG. 5, which will not be repeated here.
  • the adjustment module 802 is configured to adjust the parameters of the target service level according to the maximum delay associated with the target service level based on that any information in the related information does not meet the threshold corresponding to any information.
  • the functions performed by the adjustment module 802 may refer to the related content described in 402 in FIG. 4, or refer to the related content described in 504-510 shown in FIG. 5, which will not be repeated here.
  • the method provided by the embodiment of the application dynamically updates the queue resource allocation and data flow admission constraints based on changes in service deployment conditions (for example, current data flow parameters and remaining data flow parameters), queue status information and the number of times of reporting abnormal information, etc.
  • the device provided in FIG. 8 realizes its functions
  • only the division of the above-mentioned functional modules is used as an example for illustration.
  • the above-mentioned functions can be allocated by different functional modules according to needs, i.e.
  • the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.
  • the device and method embodiments provided in the above-mentioned embodiments belong to the same conception, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.
  • the embodiment of the present application proposes a system for adjusting the service level, which can further improve the utilization of network resources on the premise of ensuring the end-to-end delay requirement of the service.
  • the embodiment of the present application can ensure that the data flow in the network will not cause delay degradation or packet loss.
  • a QoS adjustment system based on service levels and guaranteeing end-to-end delay can improve the efficiency of network resource utilization under the premise of guaranteeing end-to-end delay of services.
  • the overall architecture of the system includes several units on the controller side and the network equipment side, as well as the related functions of each unit and the mutual interaction logic.
  • the overall workflow including the controller's global trigger QoS adjustment process, and the corresponding original information collection, adjustment trigger judgment, adjustment calculation, adjustment execution, adjustment information synchronization and other steps, which can be based on service deployment, actual arrival of data flow, and queue status
  • the changes of information, etc. realize the dynamic configuration update of the service level parameters and the dynamic switching of the service level through which the data flow passes.
  • the dynamic configuration update of the service level or the service level switching of the data stream is performed, the delay degradation of the network data stream or packet loss will not be caused.
  • the configuration update of service level parameters includes one or more of the following parameters: increase or decrease queue resource allocation, such as queue reserved bandwidth, buffering, etc.; increase or decrease data flow constraint parameters, such as data flow burst threshold and convergence Maximum average flow rate, etc.
  • the service level switching of the data stream includes: switching to other service levels of the same device and the same port; switching to other ports of the same device or the service level of other devices.
  • the network device sends related information about the target service level to the control device, where the related information includes at least one queue status information of the target service level, remaining data flow parameters, current data flow parameters, and error information reporting At least one type of information in the number of times.
  • the control device obtains the relevant information of the target service level of the first network device.
  • the control device adjusts the parameter of the target service level according to the maximum delay associated with the target service level based on that any information in the related information does not meet the threshold corresponding to the any information.
  • the functions performed by the control device refer to the related content described in 402 in FIG. 4, or refer to the related content described in 504-510 shown in FIG. 5, which will not be repeated here.
  • the control device or the network device in the foregoing embodiment may be a router or a switch.
  • the hardware structure includes but is not limited to the following two types:
  • control device or network device includes a transceiver, a processor, and a memory.
  • the transceiver of the control device is used to receive messages or data information, etc., for example, referring to the related content described in 401 shown in FIG. 4 or 501-503 in FIG.
  • the device is configured to receive information such as at least one queue status information, current data flow parameters, and reporting times of abnormal information of the target service level of the first network device reported by the first network device, and can also receive some parameters stored in the processor, such as The remaining data stream parameters.
  • the processor of the control device is configured to adjust the parameters of the target service level according to the maximum delay associated with the target service level based on any information in the related information that does not meet the threshold corresponding to the any information, for example, see FIG. 4
  • the control device or network device includes a main control board and an interface board, the main control board includes a processor and a memory, and the interface board includes a processor, a memory and an interface card.
  • the processor of the interface board is used to call the program instructions in the memory of the interface board to execute message reception and transmission.
  • the processor of the main control board is used to call the program instructions in the memory of the main control board to perform corresponding processing functions.
  • the interface card of the interface board of the control device is used to receive messages or data information, for example, see 401 shown in FIG. 4 or related content described in 501-503 of FIG.
  • the interface card of the interface board of the control device is used to receive at least one queue status information of the target service level of the first network device reported by the first network device, current data flow parameters, and reporting times of abnormal information, and can also receive the interface board Some parameters stored in the processor, such as remaining data flow parameters.
  • the processor of the main control board of the control device is used to adjust the parameters of the target service level according to the maximum delay associated with the target service level based on that any information in the related information does not meet the threshold corresponding to the any information, for example, See the related description of 402 shown in FIG. 4, or the related description of the parameters of adjusting the target service level by the control device described in 504-510 shown in FIG. 5.
  • any of the device embodiments described above are merely illustrative, where the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one unit. Locally, or it can be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines.
  • the steps of the method or algorithm described in the disclosure of the embodiments of the present application may be implemented in a hardware manner, or may be implemented in a manner in which a processor executes software instructions.
  • Software instructions can be composed of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read only memory (ROM), erasable programmable read-only memory (erasable programmable ROM (EPROM), electrically erasable programmable read-only memory (electrically erasable programmable read-only memory (EPROM, EEPROM), hard disk, mobile hard disk, optical disk, or any other form of storage medium known in the art.
  • An exemplary storage medium is coupled to the processor, so that the processor can read information from the storage medium and write information to the storage medium.
  • the storage medium may also be an integral part of the processor.
  • the processor and the storage medium may be located in the ASIC.
  • the ASIC may be located in the core network interface device.
  • the processor and the storage medium may also exist as discrete components in the core network interface device.
  • the functions described in this application can be implemented by hardware, software, firmware, or any combination thereof.
  • these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on the computer-readable medium.
  • the computer-readable medium includes a computer storage medium and a communication medium, where the communication medium includes any medium that facilitates the transfer of a computer program from one place to another.
  • the storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.
  • FIG. 11 is a schematic diagram of the hardware structure of the device 1100 for adjusting the service level according to an embodiment of the application.
  • the device 1100 for adjusting the service level shown in FIG. 11 can execute the corresponding steps in the configuration data management method provided in the embodiment shown in FIG. 3.
  • the device 1100 for adjusting the service level includes a processor 1101, a memory 1102, an interface 1103, and a bus 1104.
  • the interface 1103 may be implemented in a wireless or wired manner.
  • the interface 1103 may be a network card.
  • the aforementioned processor 1101, memory 1102, and interface 1103 are connected through a bus 1104.
  • the interface 1103 may include a transmitter and a receiver, which are used to communicate with other communication devices, and perform the related steps shown in 401 shown in FIG. 4 or shown in FIG. 5.
  • the processor 1101 is configured to execute the processing related steps shown in 402 in the embodiment shown in FIG. 4 or shown in FIG. 5.
  • the memory 1102 includes an operating system 11021 and an application program 11022, which are used to store programs, codes, or instructions. When the processor or hardware device executes these programs, codes, or instructions, the processing process of the device 1100 involved in adjusting the service level in the method embodiment can be completed. .
  • the memory 1102 may include a read-only memory (English: Read-only Memory, abbreviation: ROM) and a random access memory (English: Random Access Memory, abbreviation: RAM).
  • ROM includes basic input/output system (English: Basic Input/Output System, abbreviation: BIOS) or embedded system
  • BIOS Basic Input/Output System
  • RAM includes application programs and operating system.
  • FIG. 11 only shows a simplified design of the device 1100 for adjusting the service level.
  • the device 1100 for adjusting the service level may include any number of interfaces, processors or memories.
  • processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processing (digital signal processing, DSP), and application specific integrated circuits. ASIC), field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or any conventional processor. It is worth noting that the processor may be a processor that supports an advanced reduced instruction set machine (advanced RISC machines, ARM) architecture.
  • the foregoing memory may include a read-only memory and a random access memory, and provide instructions and data to the processor.
  • the memory may also include non-volatile random access memory.
  • the memory can also store device type information.
  • the memory may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
  • the non-volatile memory can be read-only memory (ROM), programmable read-only memory (programmable ROM, PROM), erasable programmable read-only memory (erasable PROM, EPROM), and electrically available Erase programmable read-only memory (electrically EPROM, EEPROM) or flash memory.
  • the volatile memory may be random access memory (RAM), which is used as an external cache. By way of exemplary but not limiting illustration, many forms of RAM are available.
  • static random access memory static random access memory
  • dynamic random access memory dynamic random access memory
  • DRAM dynamic random access memory
  • SDRAM synchronous dynamic random access memory
  • double data rate synchronous dynamic random access Memory double data date SDRAM, DDR SDRAM
  • enhanced synchronous dynamic random access memory enhanced SDRAM, ESDRAM
  • serial link DRAM SLDRAM
  • direct memory bus random access memory direct rambus RAM
  • the present application also provides a computer-readable storage medium, in which at least one instruction is stored, and the instruction is loaded and executed by a processor to implement any one of the above-mentioned methods for adjusting a service level. For example, the method in FIG. 4 or FIG. 5 can be performed.
  • This application provides a computer program.
  • the processor or the computer can execute the corresponding steps and/or processes in the method embodiment in FIG. 4 or FIG. 5.
  • a chip including a processor, which is used to call and execute instructions stored in the memory from a memory, so that a communication device installed with the chip can execute the method in FIG. 4 or FIG. 5.
  • Another chip including: an input interface, an output interface, a processor, and a memory.
  • the input interface, output interface, the processor, and the memory are connected through an internal connection path, and the processor is used to execute all The code in the memory, when the code is executed, the processor is used to execute the methods in the foregoing aspects.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
  • the computer instructions may be transmitted from a website, computer, server, or data center.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center integrated with one or more available media.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, and a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk).

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Environmental & Geological Engineering (AREA)
  • Quality & Reliability (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Devices For Checking Fares Or Tickets At Control Points (AREA)
  • Input Circuits Of Receivers And Coupling Of Receivers And Audio Equipment (AREA)
  • Networks Using Active Elements (AREA)

Abstract

本申请公开了调整服务等级的方法、装置、设备、系统及存储介质，该方法包括：控制设备获取网络设备的目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及异常信息上报次数中的至少一种相关信息；基于相关信息中的任一信息不满足与该任一信息对应的阈值，控制设备根据目标服务等级关联的最大时延调整目标服务等级的参数。通过调整目标服务等级的参数，使得在保障业务端到端时延的前提下，提升网络资源利用率。

Description

调整服务等级的方法、装置、设备、系统及存储介质
本申请要求于2020年01月22日提交的申请号为202010075494.5、发明名称为“动态调整服务等级的方法、设备和系统”的中国专利申请的优先权，本申请要求于2020年07月03日提交的申请号为202010635155.8、发明名称为“调整服务等级的方法、装置、设备、系统及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信领域，尤其涉及调整服务等级的方法、装置、设备、系统及存储介质。
背景技术
新兴的第五代(fifth-generation，5G)超高可靠性超低时延通信(Ultra-Reliable and Low Latency Communication，uRLLC)业务对互联网协议(internet protocol，IP)网络的服务等级协议(service level agreement，SLA)提出新的要求，在传统的可达性以及带宽保障的基础上，还需要提业务的端到端时延和丢包率保证。具体而言，这类业务需要保证报文的端到端时延不超过设定的最大时延，不产生丢包的要求。然而，基于尽力而为的传统IP网络无法提供确定性的转发业务的能力，无法提供可保障的端到端时延等SLA要求。
发明内容
本申请提供了一种调整服务等级的方法、装置、设备、系统及存储介质，以在保证业务端到端的时延要求的前提下，提升网络资源利用率。
第一方面，提供了一种调整服务等级的方法，该方法包括：控制设备获取第一网络设备的目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及异常信息的上报次数中的至少一种相关信息。控制设备基于相关信息中的任一信息不满足与任一信息对应的阈值，根据与目标服务等级关联的最大时延调整目标服务等级的参数。
其中，目标服务等级的至少一个队列状态信息包括目标服务等级的一个或多个队列状态信息。例如，每个服务等级绑定第一网络设备端口的一个队列或多个队列，其中将多个队列称为一组队列。如果目标服务等级绑定第一网络设备中的一组队列，目标服务等级的队列状态信息是指目标服务等级绑定的一组队列中的一个或多个队列的队列状态信息，或者，目标服务等级的队列状态信息是指目标服务等级绑定的一组队列的整体队列状态信息。队列状态信息包括队列状态信息包括但不限于本地队列的缓存占用、排队时延和报文计数等中的一个或多个。队列的缓存占用表示目标服务等级绑定的队列中的报文所占用的缓存大小；报文排队时延表示目标服务等级绑定的队列中的报文的排队时延，包括从报文进入缓存到报文被调度出去的时间间隔；报文计数表示目标服务等级绑定的队列中的报文的单位时间数据量或单位时间个数。
目标服务等级的剩余数据流参数包括第一网络设备的目标服务等级的剩余的能够允许的数据流的流量大小。目标服务等级的异常信息包括队列缓存占用异常信息、排队时延异常信 息、报文计数异常信息及错误信息中的至少一种。目标服务等级的当前数据流参数包括第一网络设备的目标服务等级的当前的数据流的流量大小。示例性地，错误信息是在目标服务等级的流量超过准入约束或资源阈值时，由网络设备向控制设备发送的ERROR信息。
通过调整目标服务等级的参数，使得在保障业务端到端时延的前提下，进一步提升网络资源利用率。
在一种可能的实现方式中，相关信息包括目标服务等级的至少一个队列状态信息，基于至少一个队列状态信息中的任一队列状态信息小于与任一队列状态信息对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。例如，以队列状态信息包括本地队列的缓存占用、报文排队时延、报文计数为例，若监测到目标服务等级的本地队列的缓存占用小于缓存占用阈值下限、或报文排队时延小于报文排队时延的阈值下限、或报文计数小于报文计数的阈值下限，则触发控制设备根据目标服务等级关联的最大时延调整目标服务等级。
在队列状态信息不满足对应阈值下限的情况下，由控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，从而在队列状态信息发生变化时，能够保障业务端到端时延，提升网络资源利用率。
在一种可能的实现方式中，相关信息包括目标服务等级的当前数据流参数，基于目标服务等级的当前数据流参数小于与当前数据流参数对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
在当前数据流参数小于对应的阈值下限的情况下，由控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，从而在考虑当前流量变化时，能够保障业务端到端时延，提升网络资源利用率。
在一种可能的实现方式中，相关信息包括目标服务等级的剩余数据流参数，基于目标服务等级的剩余数据流参数超过与剩余数据流参数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
在剩余数据流参数超过对应的阈值上限的情况下，由控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，从而在考虑剩余流量变化时，能够保障业务端到端时延，且进一步提升网络资源利用率。
在一种可能的实现方式中，目标服务等级的参数包括队列资源参数及数据流约束参数。控制设备确定数据流约束参数的第一更新值，在最大时延不变的情况下，根据数据流约束参数的第一更新值确定队列资源参数的第一更新值。控制设备根据数据流约束参数的第一更新值调整数据流约束参数，根据队列资源参数的第一更新值调整队列资源参数。
在一种可能的实现方式中，队列资源参数的第一更新值包括队列带宽的第一更新值及队列缓存的第一更新值。数据流约束参数的第一更新值包括数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值。在最大时延不变的情况下，控制设备确定目标服务等级对应的端口的总带宽、目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量、第一网络设备的多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度。控制设备根据目标服务等级对应的端口的总带宽、目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第一更新值。控制设备根据 多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值，确定队列缓存的第一更新值。
在一种可能的实现方式中，根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定队列带宽的第一更新值
Figure PCTCN2020118590-appb-000001
Figure PCTCN2020118590-appb-000002
在一种可能的实现方式中，根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值，按照如下公式确定队列缓存的第一更新值
Figure PCTCN2020118590-appb-000003
Figure PCTCN2020118590-appb-000004
其中，上述C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000005
为目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量，Q j为第j服务等级可转发的最大数据量，
Figure PCTCN2020118590-appb-000006
为目标服务等级在目标服务等级的参数调整前可转发的最大数据量，
Figure PCTCN2020118590-appb-000007
为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量；L max,L为低优先级队列中的最大报文长度，L n+1为非时延保障队列中的最大报文长度，L j为第j个服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000008
为数据流的突发量阈值的第一更新值，
Figure PCTCN2020118590-appb-000009
为目标服务等级的参数调整前数据流的突发量阈值，Δb i1为目标服务等级的数据流的突发量阈值的减少值，
Figure PCTCN2020118590-appb-000010
为数据流的平均速率阈值的第一更新值，
Figure PCTCN2020118590-appb-000011
为目标服务等级的参数调整前数据流的平均速率阈值，Δr i1为数据流的平均速率阈值的减少值，L i为目标服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000012
为非时延保障队列在目标服务等级的参数调整后可转发的最大数据量，n、i和j为正整数。
在一种可能的实现方式中，相关信息包括目标服务等级的至少一个队列状态信息，基于至少一个队列状态信息中的任一队列状态信息超过与任一队列状态信息对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。例如，以队列状态信息包括本地队列的缓存占用、报文排队时延、报文计数为例，若监测到目标服务等级的本地队列的缓存占用超过缓存占用的阈值上限、或报文排队时延超过报文排队时延的阈值上限、或报文计数超过报文计数的阈值上限，则触发控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
在一种可能的实现方式中，相关信息包括目标服务等级的当前数据流参数，基于目标服务等级的当前数据流参数超过与当前数据流参数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
在一种可能的实现方式中，相关信息包括目标服务等级的剩余数据流参数，基于目标服务等级的剩余数据流参数小于与剩余数据流参数对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
在一种可能的实现方式中，相关信息包括目标服务等级的异常信息的上报次数，基于目标服务等级的异常信息的上报次数超过与异常信息的上报次数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。例如，第一网络设备基于本地 资源在最大时延不变的情况下未满足目标服务等级的配置参数的更新值，向控制设备上报异常信息，该异常信息指第一网络设备的异常服务信息，例如该异常服务质量信息包括超出阈值的队列缓存占用、报文排队时延、报文计数信息及错误信息中的至少一种。
在一种可能的实现方式中，目标服务等级的参数包括队列资源参数及数据流约束参数。控制设备确定数据流约束参数的第二更新值。在最大时延不变的情况下，控制设备根据数据流约束参数的第二更新值确定队列资源参数的第二更新值。控制设备基于队列资源参数的第二更新值小于等于资源阈值，根据数据流约束参数的第二更新值调整数据流约束参数，根据队列资源参数的第二更新值调整队列资源参数。
在一种可能的实现方式中，目标服务等级的配置参数包括队列资源参数及数据流约束参数。控制设备确定数据流约束参数的第二更新值。在最大时延不变的情况下，控制设备根据数据流约束参数的第二更新值确定队列资源参数的第二更新值。控制设备基于队列资源参数的第二更新值大于资源阈值，将目标服务等级对应的目标数据流切换至其他服务等级，其他服务等级包括第一网络设备的其他服务等级或者其他设备的服务等级。
在一种可能的实现方式中，目标服务等级的配置参数包括队列资源参数及数据流约束参数。控制设备确定数据流约束参数的第二更新值。在最大时延不变的情况下，控制设备根据数据流约束参数的第二更新值确定队列资源参数的第二更新值。控制设备基于队列资源参数的第二更新值大于资源阈值，根据数据流约束参数的第二更新值确定队列资源参数的第三更新值，确定可抢占资源的服务等级的参数的更新值。可抢占资源的服务等级的参数的更新值用于使得队列资源参数的第三更新值满足约束条件。控制设备根据数据流约束参数的第二更新值调整数据流约束参数，根据队列资源参数的第三更新值调整队列资源参数，根据可抢占资源的服务等级的参数的更新值调整可抢占资源的服务等级的参数。
在一种可能的实现方式中，队列资源参数的第二更新值包括队列带宽的第二更新值及队列缓存的第二更新值，数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值。在最大时延不变的情况下，根据数据流约束参数的第二更新值确定队列资源参数的第二更新值，包括：在最大时延不变的情况下，确定目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量、第一网络设备的多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度；根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第二更新值；根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，确定队列缓存的第二更新值。
在一种可能的实现方式中，根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第二更新值，包括：根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量以及各个服务等级在目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定队列带宽的第二更新值
Figure PCTCN2020118590-appb-000013
Figure PCTCN2020118590-appb-000014
在一种可能的实现方式中，根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，确定队列缓存的第二更新值，包括：根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，按照如下公式确定队列缓存的第二更新值
Figure PCTCN2020118590-appb-000015
Figure PCTCN2020118590-appb-000016
其中，C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000017
为目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量，Q j为第j服务等级可转发的最大数据量，
Figure PCTCN2020118590-appb-000018
为目标服务等级在目标服务等级的参数调整前可转发的最大数据量，
Figure PCTCN2020118590-appb-000019
为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，L max,L为低优先级队列中的最大报文长度，L n+1为非时延保障队列中的最大报文长度，L j为第j个服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000020
为数据流的突发量阈值的第二更新值，
Figure PCTCN2020118590-appb-000021
为目标服务等级的参数调整前数据流的突发量阈值，Δb i2为数据流的突发量阈值的增加值，
Figure PCTCN2020118590-appb-000022
为数据流的平均速率阈值的第二更新值，
Figure PCTCN2020118590-appb-000023
为目标服务等级的参数调整前数据流的平均速率阈值，Δr i2为数据流的平均速率阈值的增加值，L i为目标服务等级队列中的最大报文长度，n、i和j为正整数。
在一种可能的实现方式中，数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值；根据数据流约束参数的第二更新值确定队列资源参数的第三更新值，包括：确定目标服务等级对应的端口的带宽、目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量、第一网络设备的多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度；根据目标服务等级对应的端口的带宽、目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第三更新值；根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，确定队列缓存的第三更新值；
确定可抢占资源的服务等级的参数的更新值，包括：根据目标服务等级对应的端口的带宽、可抢占资源的服务等级调整后可转发的最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定可抢占资源的服务等级的队列带宽的更新值；根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、可抢占资源的服务等级的数据流的突发量阈值的更新值及数据流的平均速率阈值的更新值，确定可抢占资源的服务等级的队列缓存的更新值。
在一种可能的实现方式中，根据目标服务等级对应的端口的带宽、目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及多个服务等级在目标服务等级的参 数调整前可转发的最大数据量，确定队列带宽的第三更新值，包括：根据目标服务等级对应的端口的带宽、目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定队列带宽的第三更新值
Figure PCTCN2020118590-appb-000024
Figure PCTCN2020118590-appb-000025
根据目标服务等级对应的端口的带宽、可抢占资源的服务等级调整后可转发的最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定可抢占资源的服务等级的队列带宽
Figure PCTCN2020118590-appb-000026
Figure PCTCN2020118590-appb-000027
在一种可能的实现方式中，根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，确定队列缓存的第三更新值，包括：根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，按照如下公式确定队列缓存的第三更新值
Figure PCTCN2020118590-appb-000028
Figure PCTCN2020118590-appb-000029
根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、可抢占资源的服务等级的数据流的突发量阈值的更新值及数据流的平均速率阈值的更新值，确定可抢占资源的服务等级的队列缓存
Figure PCTCN2020118590-appb-000030
Figure PCTCN2020118590-appb-000031
其中，C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000032
为目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量，
Figure PCTCN2020118590-appb-000033
为可抢占资源的服务等级调整后可转发的最大数据量，Q j为第j服务等级可转发的最大数据量，
Figure PCTCN2020118590-appb-000034
为目标服务等级的参数调整前可转发的最大数据量，
Figure PCTCN2020118590-appb-000035
为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，L max,L为低优先级队列中的最大报文长度，L n+1为非时延保障队列中的最大报文长度，L j为第j个服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000036
为数据流的突发量阈值的第二更新值，
Figure PCTCN2020118590-appb-000037
为目标服务等级的参数调整前数据流的突发量阈值，Δb i2为数据流的突发量阈值的增加值，
Figure PCTCN2020118590-appb-000038
为数据流的平均速率阈值的第二更新值，
Figure PCTCN2020118590-appb-000039
为目标服务等级的参数调整前数据流的平均速率阈值，Δr i2为数据流的平均速率阈值的增加值，L i为目标服务等级队列中的最大报文长度，L k为可抢占资源的服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000040
为可抢占资源的服务等级 调整后可转发的最大数据量，
Figure PCTCN2020118590-appb-000041
为非时延保障队列在目标服务等级的参数调整后可转发的最大数据量，
Figure PCTCN2020118590-appb-000042
为可抢占资源的服务等级的数据流的平均速率阈值的更新值，
Figure PCTCN2020118590-appb-000043
为可抢占资源的服务等级的数据流的突发量阈值的更新值，n、i和j为正整数。
在一种可能的实现方式中，第一网络设备调整目标服务等级的参数之后，还包括：控制设备将调整的目标服务等级的参数的更新值发送给第一网络设备，指示第一网络设备按照目标服务等级的参数的更新值调整目标服务等级的参数。
在一种可能的实现方式中，控制设备获取第一网络设备的目标服务等级的相关信息包括：所述控制设备在预设周期内接收所述第一网络设备发送的目标服务等级的相关信息。
第二方面，提供了一种调整服务等级的装置，装置包括：
获取模块用于获取第一网络设备的目标服务等级的相关信息，所述相关信息包括所述目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及异常信息的上报次数中的至少一种信息。调整模块用于基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
在一种可能的实现方式中，所述相关信息包括所述目标服务等级的至少一个队列状态信息。所述调整模块用于基于所述至少一个队列状态信息中的任一队列状态信息小于与所述任一队列状态信息对应的阈值下限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
在一种可能的实现方式中，所述相关信息包括所述目标服务等级的当前数据流参数。
所述调整模块用于基于所述目标服务等级的当前数据流参数小于与所述当前数据流参数对应的阈值下限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
在一种可能的实现方式中，所述相关信息包括所述目标服务等级的剩余数据流参数。
所述调整模块用于基于所述目标服务等级的剩余数据流参数超过与所述剩余数据流参数对应的阈值上限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
在一种可能的实现方式中，所述目标服务等级的参数包括队列资源参数及数据流约束参数。所述调整模块用于确定所述数据流约束参数的第一更新值；在所述最大时延不变的情况下，根据所述数据流约束参数的第一更新值确定所述队列资源参数的第一更新值；根据所述数据流约束参数的第一更新值调整所述数据流约束参数，根据所述队列资源参数的第一更新值调整所述队列资源参数。
在一种可能的实现方式中，所述队列资源参数的第一更新值包括队列带宽的第一更新值及队列缓存的第一更新值，所述数据流约束参数的第一更新值包括数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值；
所述调整模块用于在所述最大时延不变的情况下，确定所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第一最大数据量、所述第一网络设备的多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度；
根据所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第一更新值。
根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述 多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第一更新值及所述数据流的平均速率阈值的第一更新值，确定所述队列缓存的第一更新值。
在一种可能的实现方式中，所述调整模块，用于根据所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定所述队列带宽的第一更新值
Figure PCTCN2020118590-appb-000044
Figure PCTCN2020118590-appb-000045
在一种可能的实现方式中，所述调整模块，用于根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第一更新值及所述数据流的平均速率阈值的第一更新值，按照如下公式确定所述队列缓存的第一更新值
Figure PCTCN2020118590-appb-000046
Figure PCTCN2020118590-appb-000047
在一种可能的实现方式中，所述相关信息包括所述目标服务等级的至少一个队列状态信息；所述调整模块用于基于所述至少一个队列状态信息中的任一队列状态信息超过与所述任一队列状态信息对应的阈值上限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
在一种可能的实现方式中，所述相关信息包括所述目标服务等级的当前数据流参数；
所述调整模块用于基于所述目标服务等级的当前数据流参数超过与所述当前数据流参数对应的阈值上限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
在一种可能的实现方式中，所述相关信息包括所述目标服务等级的剩余数据流参数。
所述调整模块用于基于所述目标服务等级的剩余数据流参数小于与所述剩余数据流参数对应的阈值下限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
在一种可能的实现方式中，所述相关信息包括所述目标服务等级的异常信息的上报次数。
所述调整模块用于基于所述目标服务等级的异常信息的上报次数超过与所述异常信息的上报次数对应的阈值上限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
在一种可能的实现方式中，所述目标服务等级的参数包括队列资源参数及数据流约束参数；所述调整模块用于确定所述数据流约束参数的第二更新值。
在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值；
基于所述队列资源参数的第二更新值小于等于资源阈值，根据所述数据流约束参数的第二更新值调整所述数据流约束参数，根据所述队列资源参数的第二更新值调整所述队列资源参数。
在一种可能的实现方式中，所述目标服务等级的配置参数包括队列资源参数及数据流约束参数。所述调整模块用于确定所述数据流约束参数的第二更新值。
在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值；
基于所述队列资源参数的第二更新值大于资源阈值，将所述目标服务等级对应的目标数 据流切换至其他服务等级，所述其他服务等级包括所述第一网络设备的其他服务等级或者其他设备的服务等级。
在一种可能的实现方式中，所述目标服务等级的配置参数包括队列资源参数及数据流约束参数。所述调整模块用于确定所述数据流约束参数的第二更新值；在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值。
基于所述队列资源参数的第二更新值大于资源阈值，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第三更新值，确定可抢占资源的服务等级的参数的更新值，所述可抢占资源的服务等级的参数的更新值用于使得所述队列资源参数的第三更新值满足约束条件。
根据所述数据流约束参数的第二更新值调整所述数据流约束参数，根据所述队列资源参数的第三更新值调整所述队列资源参数，根据所述可抢占资源的服务等级的参数的更新值调整所述可抢占资源的服务等级的参数。
在一种可能的实现方式中，所述队列资源参数的第二更新值包括队列带宽的第二更新值及队列缓存的第二更新值，所述数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值。
所述调整模块，用于在所述最大时延不变的情况下，确定所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量、所述第一网络设备的多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度。
根据所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第二更新值。
根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，确定所述队列缓存的第二更新值。
在一种可能的实现方式中，所述调整模块，用于根据所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量以及所述各个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定所述队列带宽的第二更新值
Figure PCTCN2020118590-appb-000048
Figure PCTCN2020118590-appb-000049
在一种可能的实现方式中，所述调整模块，用于根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，按照如下公式确定所述队列缓存的第二更新值
Figure PCTCN2020118590-appb-000050
Figure PCTCN2020118590-appb-000051
在一种可能的实现方式中，所述数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值。
所述调整模块用于确定所述目标服务等级对应的端口的带宽、所述目标服务等级在所述 目标服务等级的参数调整后可转发的第三最大数据量、所述第一网络设备的多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度。
根据所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第三更新值。
根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，确定所述队列缓存的第三更新值。
所述调整模块，用于根据所述目标服务等级对应的端口的带宽、所述可抢占资源的服务等级调整后可转发的最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述可抢占资源的服务等级的队列带宽的更新值。
根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述可抢占资源的服务等级的数据流的突发量阈值的更新值及数据流的平均速率阈值的更新值，确定所述可抢占资源的服务等级的队列缓存的更新值。
在一种可能的实现方式中，所述调整模块，用于根据所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定所述队列带宽的第三更新值
Figure PCTCN2020118590-appb-000052
Figure PCTCN2020118590-appb-000053
根据所述目标服务等级对应的端口的带宽、所述可抢占资源的服务等级调整后可转发的最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述可抢占资源的服务等级的队列带宽
Figure PCTCN2020118590-appb-000054
Figure PCTCN2020118590-appb-000055
在一种可能的实现方式中，所述调整模块，用于根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，按照如下公式确定所述队列缓存的第三更新值
Figure PCTCN2020118590-appb-000056
Figure PCTCN2020118590-appb-000057
根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述可抢占资源的服务等级的数据流的突发量阈值的更新值及数据 流的平均速率阈值的更新值，确定所述可抢占资源的服务等级的队列缓存
Figure PCTCN2020118590-appb-000058
Figure PCTCN2020118590-appb-000059
在一种可能的实现方式中，所述装置，还包括：
发送模块用于将调整的所述目标服务等级的参数的更新值发送给所述第一网络设备，指示所述第一网络设备按照所述目标服务等级的参数的更新值调整所述目标服务等级的参数。
在一种可能的实现方式中，所述获取模块用于在预设周期内接收所述第一网络设备发送的目标服务等级的相关信息。
需要说明的是，上述第二方面及任一实现方式中涉及的公式中的参数的含义，可参见第一方面及第一方面的相关描述，此处不再一一赘述。
第三方面，提供了一种网络设备，执行第一方面或第一方面任一实现方式中的方法，例如，该网络设备包括用于执行第一方面实现方式中的方法的单元。
第四方面，提供了一种计算机可读存储介质，计算机可读存储介质包括计算机程序或指令，当所述计算机程序或指令在计算机上执行时，使得所述计算机执行如上第一方面及任一所述的方法。
第五方面，提供了一种网络设备，该网络设备包括存储器及处理器，所述存储器中存储有计算机程序或至少一条指令，所述计算机程序或至少一条指令由所述处理器加载并执行，以实现如上第一方面及任一所述的方法。
第六方面，提供了一种通信装置，该装置包括：收发器、存储器和处理器。其中，该收发器、该存储器和该处理器通过内部连接通路互相通信，该存储器用于存储计算机程序或指令，该处理器用于执行该存储器存储的计算机程序或指令，以控制收发器接收信号，并控制收发器发送信号，并且当该处理器执行该存储器存储的指令时，使得该处理器执行第一方面或第一方面的任一种可能的实施方式中的方法。
作为一种示例性实施例，所述处理器为一个或多个，所述存储器为一个或多个。
作为一种示例性实施例，所述存储器可以与所述处理器集成在一起，或者所述存储器与处理器分离设置。
在具体实现过程中，存储器可以为非瞬时性(non-transitory)存储器，例如只读存储器(read only memory，ROM)，其可以与处理器集成在同一块芯片上，也可以分别设置在不同的芯片上，本申请实施例对存储器的类型以及存储器与处理器的设置方式不做限定。
第七方面，提供了一种计算机程序(产品)，所述计算机程序(产品)包括：计算机程序代码，当所述计算机程序代码被计算机运行时，使得所述计算机执行上述第一方面或第一方面的任一种可能的实施方式中的方法。
第八方面，提供了一种芯片，包括处理器，用于从存储器中调用并运行所述存储器中存储的计算机程序或指令，使得安装有所述芯片的通信设备执行上述第一方面或第一方面的任一种可能的实施方式中的方法。
第九方面，提供另一种芯片，包括：输入接口、输出接口、处理器和存储器，所述输入接口、输出接口、所述处理器以及所述存储器之间通过内部连接通路相连，所述处理器用于执行所述存储器中的代码，当所述代码被执行时，所述处理器用于执行上述第一方面或第一方面的任一种可能的实施方式中的方法。
第十方面，提供一种调整服务等级的系统，包括上述第二方面或第二方面的任一种可能 的实现方式的控制设备和第一网络设备，第一网络设备用于向控制设备发送目标服务等级的相关信息，所述目标服务等级的相关信息包括所述目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及异常信息的上报次数中的至少一种信息。
附图说明
图1为本申请实施例提供的一种调整服务等级的场景示意图；
图2为本申请实施例提供的一种调整服务等级的场景示意图；
图3为本申请实施例提供的一种调整服务等级的系统架构示意图；
图4为本申请实施例提供的一种调整服务等级的方法流程示意图；
图5为本申请实施例提供的一种调整服务等级的方法流程示意图；
图6为本申请实施例提供的一种调整服务等级的出端口队列示意图；
图7为本申请实施例提供的一种调整服务等级的网络演算示意图；
图8为本申请实施例提供的一种调整服务等级的装置结构示意图；
图9为本申请实施例提供的一种网络设备的结构示意图；
图10为本申请实施例提供的一种网络设备的结构示意图；
图11为本申请实施例提供的一种调整服务等级的设备的结构示意图。
具体实施方式
为了保障业务数据流的端到端时延需求，传统的网络服务质量(quality of service,QoS)保障的实现包括集成服务(integrated service,IntServ)模型和差分服务(Differentiated Services,DiffServ)模型来实现。其中，IntServ模型依赖网络中的网络设备进行逐流的状态维护、资源预留和准入控制。DiffServ模型不要求网络设备对业务数据流进行逐流维护，而是根据报文携带的差分服务类型，对报文实施基于优先级的逐跳行为转发操作。
然而，IntServ模型能够保障业务端到端的QoS，但其扩展性较差、难以大范围部署。DiffServ模型扩展性虽然好，但受限于网络设备优先级策略的相对性特点，无法提供确定的转发能力，进而无法严格保证业务数据流的端到端时延需求。因此，两种模型都无法在当前的大规模IP网络中实现端到端的时延保障。并且，当前基于SLA保障端到端时延的QoS技术，没有考虑在业务需求、实时数据流和网络状态等发生变化时，服务等级的QoS动态调整技术。如果保持静态的QoS设置不变，无法保证在相关条件变化时仍然最优。
因此，本申请实施例提出了一种调整服务等级的方法，可以根据业务部署情况(例如当前数据流参数和剩余数据流参数)、队列状态信息及异常信息的上报次数等的变化，动态更新队列资源分配和数据流准入约束等服务等级的参数配置，以及对数据流经过的服务等级进行动态切换，在保障业务端到端时延需求的前提下，进一步提升网络资源利用率。
下面以图1所示的场景示意图为例对调整服务等级的方法进行介绍。图1中，包括两个网络设备和一个控制设备，网络设备分别为第一网络设备和第二网络设备，第一网络设备和第二网络设备的发送业务数据流的出端口分别维护一个或多个服务等级，并为服务等级所绑定的队列分配相应的带宽、缓存等资源。第一网络设备和第二网络设备为传输有时延需求的数据流的路径上的网络设备。第一网络设备和第二网络设备分别将自身的服务等级及关联的最大时延、置信系数、数据流准入约束等信息中的一个或多个上报给控制设备。控制器获取业 务数据流签约承诺的服务参数，包括最大允许突发、最大允许平均速率等信息中的一个或多个。控制设备基于获取的上述信息进行业务部署，包括选择转发路径、选择路径上每跳网络设备要进入的服务等级，以保证满足业务数据流的端到端时延要求，同时确保所选服务等级满足各自的数据流准入约束。在一种示例中，控制设备将转发路径、每跳服务等级选择等信息下发至网络设备，使业务数据流按照指定路径和服务等级进行转发操作。
控制设备既可以如图1所示，为独立的物理设备，即物理上独立于第一网络设备和第二网络设备；也可以是一个功能单元，部署在第一网络设备或第二网络设备上。只要控制设备具备逻辑上相应的管理和控制功能即可，本申请对控制设备的存在形式不做限制。第一网络设备和第二网络设备可以为硬件或软硬结合的形式，是一个独立的网络设备，例如交换机、路由器等具有转发功能的网络设备，用于对网络中的数据流进行接收和发送。第一网络设备和第二网络设备也可以是软件的形式，是网络中其他网络设备上的一个功能模块或多个功能模块的组合，可以根据具体场景需求进行选择和设计。
又例如，以图2所示的网络场景为例，该网络场景包括控制器101和若干网络设备，其中，若干网络设备包括第一网络设备102、第二网络设备103以及第三网络设备104等。需要说明的是，若干网络设备的功能均可参见第一网络设备102的描述。示例性地，第一网络设备102是一个或多个具有报文转发功能的路由器或交换机。控制设备101可由具有路径计算、资源分配等功能的服务器、网络设备或其他软件、软硬结合等形式，本申请实施例仅以针对控制设备101为控制器进行描述为例。
针对图2所示的网络场景，在第一网络设备102转发数据流的端口设置一个或多个服务等级，每个服务等级提供相应的转发服务能力。或者说，第一网络设备102所提供的确定性的转发服务能力定义为第一网络设备102的服务等级。每个服务等级绑定第一网络设备102中的一个队列或一组队列。每个服务等级关联一个时延阈值，例如最大时延。该时延阈值表示输入与服务等级绑定的队列的数据流在第一网络设备102转发过程中所允许的时延，该时延阈值包括第一网络设备的排队时延、处理时延和发送时延等时延中的一个或多个。其中，排队时延指报文进入队列排队所消耗的时间。例如，排队时延是指报文进入下行流量管理器的队列排队所消耗的时间。发送时延指网络设备发送报文所需要的时间，也即从发送报文的第一个比特，到该报文的最后一个比特发送完毕所需的时间。处理时延指网络设备在收到报文时进行报文头解析、差错校验、路由查找等操作消耗的时间。在一种示例中，时延阈值为第一网络设备的排队时延；在另一种示例中，时延阈值为第一网络设备的排队时延、处理时延和发送时延之和。
可选地，每个服务等级还可关联一个置信系数，或称为可靠性概率，该置信系数表示数据流经过该服务等级时产生的时延不超过上述时延阈值的概率。默认情况下该置信系数可设为100％或1，表示产生的时延严格不超过该时延阈值。
为了保证服务等级所承诺的最大时延需求，一方面需要为该服务等级所绑定的队列分配资源，包括但不限于队列带宽、队列缓存等。示例性地，在轮询调度机制下，队列带宽即为该队列权重占总权重的比例乘以出端口的带宽。另一方面，也要对输入队列的数据流设置约束条件，包括但不限于进入队列的数据流的突发量阈值、数据流的平均速率阈值等。数据流的突发量阈值表示为数据流在设定时间内允许到达的最大数据量，数据流的平均速率阈值表示为数据流在设定时间内的平均速率所允许的最大值。在给定队列资源和数据流约束条件前 提下，可以确保数据流经过该服务等级时，能够满足特定最大时延要求。
在外部条件变化时，一方面，可以考虑调整更新服务等级的队列资源分配和数据流准入约束等参数的配置，以提升资源利用率。
例如，对于第一网络设备102的目标服务等级，如果判断进入该目标服务等级的数据流相对较多，达到数据流准入约束上限，导致预先分配给该目标服务等级的带宽、缓存等资源相对不足的情况下，则在保持该目标服务等级关联的最大时延不变的前提下，增加对该目标服务等级的队列资源分配，例如增加带宽、缓存等。还可以进一步调整数据流准入约束，例如增加数据流的突发量阈值、数据流的平均速率阈值等。这样可以消除该目标服务等级的瓶颈，容纳更多的数据流，从而提升资源利用率。
又例如，对于第一网络设备102的目标服务等级，如果一段时间内经过该目标服务等级的数据流相对较少，例如远未达到数据流准入约束上限，导致预先分配给该目标服务等级的带宽、缓存等资源相对空闲。可以考虑在保持该目标服务等级的最大时延不变的前提下，减少对该目标服务等级的资源分配，例如减少带宽、缓存等。还可以进一步调整数据流准入约束。在第一网络设备的本地资源总量一定的前提下，该目标服务等级释放出的资源，转移给该第一网络设备的其他服务等级，增加其他服务等级的资源分配，有利于这些服务等级容纳更多数据流需求，最终提升整体资源利用率。
另一方面，可以考虑调整已部署数据流沿路的服务等级，也即对数据流沿路进入的服务等级进行设备内部乃至跨设备的切换，以提升资源利用率。
例如，对于已部署的数据流，若该沿路经过的目标服务等级接近或达到数据流准入约束上界，影响更多数据流的进一步部署，可以考虑在确保仍然能满足业务的端到端时延需求的前提下，将该数据流从目标服务等级切换至本设备或其他设备的相对更空闲的服务等级之中，以释放原位置的资源供其他数据流部署，从而提升整体资源利用率。
以本申请实施例的系统架构如图3所示，该系统架构由控制设备101和第一网络设备102的若干单元模块组成，本申请实施例以图3中的第一网络设备102进行示例性说明。示例性地，控制设备101包括但不限于全局业务部署单元111、全局队列监控分析单元112、全局调整触发单元113、全局配置更新计算单元114、全局服务等级切换单元115、全局服务等级维护单元116。第一网络设备102包括但不限于本地队列监控单元121、本地调整触发单元122、本地配置更新计算单元123、本地配置更新执行单元124、本地异常信息上报单元125。控制设备101和第一网络设备102的若干单元模块的功能参见图5所示方法流程的相关描述。在一种示例中，控制设备101和第一网络设备102部署在同一网络设备中，该网络设备具备控制设备101和第一网络设备102所示的功能模块或单元。
本申请实施例提供的方法，包括但不限于在全局触发QoS调整流程下，控制器根据全局业务数据流的准入、全局队列监控分析或本地异常信息上报决定是否触发全局的QoS调整；若决定触发QoS调整，则控制器进行服务等级的更新、或者对已部署数据流的服务等级进行切换。
通过控制设备101全局触发完整的调整流程，能够保证在进行服务等级的配置更新或数据流的服务等级切换时，不会引起网络数据流的时延劣化或丢包。示例性地，本申请实施例提供的方法可适用于基于报文统计复用的IP网络，有助于提升基于服务等级保障端到端时延方案的适用性和有效性。
下面以控制设备全局触发服务等级的动态调整流程为例，对本申请实施例提供的调整服务等级的方法流程进行说明。如图4所示，该方法流程包括如下几个过程。其中，控制设备与网络设备之间的通信通过NETCONF实现。
401，控制设备获取第一网络设备的目标服务等级的相关信息，相关信息包括目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及异常信息的上报次数中的至少一种信息。
控制设备监控各个网络设备的服务等级，获取第一网络设备的目标服务等级的相关信息，其中，目标服务等级可以理解为被监控的服务等级。在一种示例中，该相关信息包括目标服务等级至少一个队列状态信息；在另一种示例中，该相关信息包括目标服务等级的剩余数据流参数；在另一种示例中，该相关信息包括目标服务等级的当前数据流参数；在另一种示例中，该相关信息包括目标服务等级的异常信息的上报次数。
例如，第一网络设备在设定周期内，将自身的服务等级的队列状态信息上报给控制设备，控制设备获取到第一网络设备的目标服务等级的队列状态信息。
又例如，控制设备本地维护各个网络设备的服务等级的剩余数据流参数以及当前数据流参数，从而获取到第一网络设备的剩余数据流参数以及当前数据流参数。示例性地，当前数据流参数包括但不限于当前数据流的突发量以及平均速率。剩余数据流参数包括但不限于剩余数据流的突发量以及平均速率。当前数据流参数可由网络设备向控制设备上报，剩余数据流参数由控制设备根据当前数据流参数得到。
又例如，第一网络设备基于本地资源在最大时延不变的情况下未满足目标服务等级的配置参数的更新值，向控制设备上报异常信息，该异常信息包括超出阈值的队列缓存占用、报文排队时延、报文计数信息及错误信息中的至少一种。由此，控制设备根据第一网络设备上报的异常服务质量信息统计得到异常信息的上报次数。
402，基于相关信息中的任一信息不满足与该任一信息对应的阈值，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
情况一：目标服务等级的相关信息包括目标服务等级的至少一个队列状态信息。
基于相关信息中的任一信息不满足与任一信息对应的阈值，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：
情况一A，基于至少一个队列状态信息中的任一队列状态信息小于与任一队列状态信息对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。在示例性实施例中，控制设备在所控制的各个网络设备的端口设置一个或多个服务等级，控制设备实时监控各个网络设备的各个服务等级的队列状态的指标，从而获取到第一网络设备的目标服务等级的至少一个队列状态信息。
示例性地，每个服务等级绑定第一网络设备中的一个队列或多个队列，其中将多个队列称为一组队列。如果目标服务等级绑定第一网络设备中的一组队列，目标服务等级的队列状态信息是指目标服务等级绑定的一组队列中的一个或多个队列的队列状态信息，或者，目标服务等级的队列状态信息是指目标服务等级绑定的一组队列的整体队列状态信息。
该目标服务等级的队列状态信息包括但不限于本地队列的缓存占用、排队时延和报文计数等至少一个队列状态信息。其中，队列的缓存占用表示目标服务等级绑定的队列中的报文所占用的缓存大小；报文排队时延表示目标服务等级绑定的队列中的报文的排队时延，包括 从报文进入缓存到报文被调度出去的时间间隔；报文计数表示目标服务等级绑定的队列中的报文的单位时间数据量或单位时间个数。
在一种示例中，目标服务等级的队列状态信息指目标服务等级绑定的一组队列的整体队列状态信息时，针对队列的缓存占用，该整体队列状态信息是指绑定的一组队列的缓存占用之和，也即是将绑定的一组队列中的各个队列的缓存占用进行累加，得到的结果作为整体队列状态信息。针对报文排队时延，该整体队列状态信息是指绑定的一组队列中，各个队列的报文排队时延中的最大值，或者各个队列的报文排队时延的平均值。针对报文计数，该整体队列状态信息是指绑定的一组队列的报文计数之和，也即是将绑定的一组队列中的各个队列的报文计数进行累加，得到的结果作为整体队列状态信息。
以队列状态信息包括本地队列的缓存占用、报文排队时延、报文计数为例，若监测到目标服务等级的本地队列的缓存占用小于缓存占用阈值下限、或报文排队时延小于报文排队时延的阈值下限、或报文计数小于报文计数的阈值下限，则触发控制设备根据目标服务等级关联的最大时延调整目标服务等级。
需要说明的是，针对目标服务等级绑定第一网络设备中的一组队列，目标服务等级的队列状态信息是指目标服务等级绑定的一组队列中的每个队列的队列状态信息，该目标服务等级的各个队列状态信息对应的阈值下限可根据单个队列来设置，一组队列中的各个队列的队列状态信息所对应的阈值下限可以相同，也可以不相同。针对目标服务等级的队列状态信息是指目标服务等级绑定的一组队列的队列状态信息的情况，该目标服务等级的各个队列状态信息对应的阈值下限可根据一组队列中的队列数量来设置。除此之外，还可以采用其他方式，关于与每种队列状态信息对应的阈值下限，本申请实施例不进行限定，例如，缓存占用阈值下限、报文排队时延的阈值下限及报文计数的阈值下限可基于经验设置，还可根据应用场景进行设置。
无论是哪种队列状态信息超过对应的阈值下限，基于至少一个队列状态信息中的任一队列状态信息小于与任一队列状态信息对应的阈值下限，均可触发控制设备调整目标服务等级。
此外，本申请实施例提供的方法中，基于至少一个队列状态信息中的任一队列状态信息小于与任一队列状态信息对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级，包括：基于至少一个队列状态信息中的任一队列状态信息在第一参考时间段内小于与该任一队列状态信息对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级。通过在第一参考时间段内监测到任一队列状态信息均小于与该任一队列状态信息对应的阈值下限再来触发根据与目标服务等级关联的最大时延调整目标服务等级，从而能够防止因队列状态信息反复出现不小于与队列状态信息对应的阈值下限而反复调整目标服务等级的情况，进一步提高全局资源使用的稳定性及准确性。
情况一B，基于至少一个队列状态信息中的任一队列状态信息超过与任一队列状态信息对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
示例性地，第一网络设备的每个服务等级绑定第一网络设备中的一个队列或一组队列。如果目标服务等级绑定第一网络设备中的一组队列，目标服务等级的队列状态信息是指目标服务等级绑定的一组队列中的每个队列的队列状态信息，或者，目标服务等级的队列状态信息是指目标服务等级绑定的一组队列的队列状态信息。该目标服务等级的队列状态信息包括但不限于本地队列的缓存占用、报文排队时延、报文计数等至少一个队列状态信息。
以队列状态信息包括本地队列的缓存占用、报文排队时延、报文计数为例，若监测到目标服务等级的本地队列的缓存占用超过缓存占用的阈值上限、或报文排队时延超过报文排队时延的阈值上限、或报文计数超过报文计数的阈值上限，则触发控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
此外，本申请实施例提供的方法中，基于至少一个队列状态信息中的任一队列状态信息超过与任一队列状态信息对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级，包括：基于至少一个队列状态信息中的任一队列状态信息在第二参考时间段内超过与任一队列状态信息对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。通过在第二参考时间段内监测到任一队列状态信息均超过与该任一队列状态信息对应的阈值上限再来触发根据与目标服务等级关联的最大时延调整目标服务等级，从而能够防止因队列状态信息反复出现小于等于队列状态信息对应的阈值下限而反复调整目标服务等级的参数情况，进一步提高全局资源使用的稳定性及准确性。
情况二：目标服务等级的相关信息包括目标服务等级的当前数据流参数。
基于相关信息中的任一信息不满足与该任一信息对应的阈值，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：
情况二A：基于目标服务等级的当前数据流参数小于与当前数据流参数对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
本申请实施例提供的方法中，基于目标服务等级的当前数据流参数小于与当前数据流参数对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：基于目标服务等级的当前数据流参数在第三参考时间段内均小于与当前数据流参数对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。通过在第三参考时间段内监测到目标服务等级的当前数据流参数均小于与该当前数据流参数对应的阈值下限再来触发根据与目标服务等级关联的最大时延调整目标服务等级的参数，从而能够防止因目标服务等级的当前数据流参数反复出现不小于与当前数据流参数对应的阈值下限而反复调整目标服务等级的参数的情况，进一步提高资源使用的稳定性及准确性。
情况二B：基于目标服务等级的当前数据流参数超过与当前数据流参数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
本申请实施例提供的方法中，基于目标服务等级的当前数据流参数超过与当前数据流参数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：基于目标服务等级的当前数据流参数在第四参考时间段内均超过与当前数据流参数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。通过在第四参考时间段内监测到目标服务等级的当前数据流参数均超过与该当前数据流参数对应的阈值上限再来触发根据与目标服务等级关联的最大时延调整目标服务等级的参数，从而能够防止因目标服务等级的当前数据流参数反复出现小于等于与当前数据流参数对应的阈值上限而反复调整目标服务等级的参数的情况，进一步提高资源使用的稳定性及准确性。
情况三：目标服务等级的相关信息包括目标服务等级的剩余数据流参数。
基于相关信息中的任一信息不满足与该任一信息对应的阈值，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：
情况三A：基于目标服务等级的剩余数据流参数超过与剩余数据流参数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
本申请实施例提供的方法中，基于目标服务等级的剩余数据流参数超过与剩余数据流参数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：基于目标服务等级的剩余数据流参数在第五参考时间段内均超过与剩余数据流参数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级。通过在第五参考时间段内监测到目标服务等级的剩余数据流参数均超过与该目标服务等级的剩余数据流参数对应的阈值上限再来触发根据与目标服务等级关联的最大时延调整目标服务等级，从而能够防止因目标服务等级的剩余数据流参数反复出现小于等于与目标服务等级的剩余数据流参数对应的阈值上限而反复调整目标服务等级的情况，进一步提高资源使用的稳定性及准确性。
情况三B：基于目标服务等级的剩余数据流参数小于与剩余数据流参数对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
本申请实施例提供的方法中，基于目标服务等级的剩余数据流参数小于与剩余数据流参数对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：基于目标服务等级的剩余数据流参数在第六参考时间段内均小于与剩余数据流参数对应的阈值下限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级。通过在第六参考时间段内监测到目标服务等级的剩余数据流参数均小于与该目标服务等级的剩余数据流参数对应的阈值下限再来触发根据与目标服务等级关联的最大时延调整目标服务等级的参数，从而能够防止因目标服务等级的剩余数据流参数反复出现不小于与目标服务等级的剩余数据流参数对应的阈值下限而反复调整目标服务等级的参数的情况，进一步提高资源使用的稳定性及准确性。
情况四：目标服务等级的相关信息包括目标服务等级的异常信息的上报次数。
基于相关信息中的任一信息不满足与任一信息对应的阈值，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：基于目标服务等级的异常信息的上报次数超过与异常信息的上报次数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数。
本申请实施例提供的方法中，基于目标服务等级的异常信息的上报次数超过与异常信息的上报次数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：基于目标服务等级的异常信息的上报次数在第七时间段内超过与异常信息的上报次数对应的阈值上限，控制设备根据与目标服务等级关联的最大时延调整目标服务等级。通过在第七参考时间段内监测到目标服务等级的异常信息的上报次数超过与异常信息的上报次数对应的阈值上限再来触发根据与目标服务等级关联的最大时延调整目标服务等级的参数，从而能够防止因时间过长才出现目标服务等级的异常信息的上报次数超过与异常信息的上报次数对应的阈值上限而去调整目标服务等级的参数，使得调整的时机与实际情况不符的情况，从而进一步提高资源使用的稳定性及准确性。
针对上述情况一A、情况二A、情况三A，目标服务等级的参数包括队列资源参数及数据流约束参数。其中，根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括：确定目标服务等级的数据流约束参数的第一更新值；在最大时延不变的情况下，根据目标服 务等级的数据流约束参数的第一更新值确定目标服务等级的队列资源参数的第一更新值；然后，根据数据流约束参数的第一更新值调整之前的数据流约束参数，根据队列资源参数的第一更新值调整之前的队列资源参数。
在示例性实施例中，队列资源参数的第一更新值包括队列带宽的第一更新值及队列缓存的第一更新值，数据流约束参数的第一更新值包括数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值。
在最大时延不变的情况下，根据数据流约束参数的第一更新值确定队列资源参数的第一更新值，包括：在最大时延不变的情况下，确定目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量、第一网络设备的多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度；根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第一更新值；根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值，确定队列缓存的第一更新值。
示例性地，根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第一更新值，包括：根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定队列带宽的第一更新值
Figure PCTCN2020118590-appb-000060
Figure PCTCN2020118590-appb-000061
其中，C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000062
为目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量，Q j为第j服务等级可转发的最大数据量，
Figure PCTCN2020118590-appb-000063
为目标服务等级在目标服务等级的参数调整前可转发的最大数据量，
Figure PCTCN2020118590-appb-000064
为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，n为大于1的正整数。
在示例性实施例中，
Figure PCTCN2020118590-appb-000065
和
Figure PCTCN2020118590-appb-000066
包括但不限于是在本次调整之前，由控制设备上的配置得到的数值，例如，控制设备在本次调整目标服务等级的参数之前配置了目标服务等级可转发的最大数据量，则针对本次调整，该
Figure PCTCN2020118590-appb-000067
即为目标服务等级在目标服务等级的参数调整前可转发的最大数据量。同理，控制设备在本次调整目标服务等级的参数之前配置了非时延保障队列可转发的最大数据量，则针对本次调整，该
Figure PCTCN2020118590-appb-000068
即为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量。
此外，
Figure PCTCN2020118590-appb-000069
和
Figure PCTCN2020118590-appb-000070
包括但不限于是上次调整目标服务等级的参数之后得到的值。在上次调整目标服务等级的参数之后，控制设备将调整后的
Figure PCTCN2020118590-appb-000071
和
Figure PCTCN2020118590-appb-000072
进行了存储，则针对本次调整，获取已经存储的
Figure PCTCN2020118590-appb-000073
作为目标服务等级在目标服务等级的参数调整前可转发的最大数据量，获取已经存储的
Figure PCTCN2020118590-appb-000074
作为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量。
针对Q j，由于目标服务等级的参数进行调整，并不会涉及第j服务等级的参数发生变化，因而第j服务等级可转发的最大数据量可根据控制设备的配置得到。或者，在本次调整目标服务等级的参数之前，控制设备对第j服务等级的参数进行了调整，控制设备也可对调整后的值 进行存储。则针对本次调整目标服务等级的参数，可直接获取已存储的Q j作为本次目标服务等级的参数调整时使用的第j服务等级可转发的最大数据量。
示例性地，根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值，确定队列缓存的第一更新值，包括：
根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值，按照如下公式确定队列缓存的第一更新值
Figure PCTCN2020118590-appb-000075
Figure PCTCN2020118590-appb-000076
其中，L max,L为低优先级队列中的最大报文长度，C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000077
为目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量，Q j为第j服务等级可转发的最大数据量，L n+1为非时延保障队列中的最大报文长度，L j为第j个服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000078
为数据流的突发量阈值的第一更新值，
Figure PCTCN2020118590-appb-000079
为目标服务等级的参数调整前数据流的突发量阈值，Δb i1为目标服务等级的数据流的突发量阈值的减少值，
Figure PCTCN2020118590-appb-000080
为数据流的平均速率阈值的第一更新值，
Figure PCTCN2020118590-appb-000081
为目标服务等级的参数调整前数据流的平均速率阈值，Δr i1为数据流的平均速率阈值的减少值，L i为目标服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000082
为非时延保障队列在目标服务等级的参数调整后可转发的最大数据量，n为大于1的正整数。i和j均为正整数。
在示例性实施例中，L n+1、L j、L i、L max,L和C可在控制设备上预先存储，例如，控制设备为目标服务等级对应的端口分配带宽之后，将L n+1、L j、L i、L max,L和C等进行存储。则控制设备直接获取存储的L n+1、L j、L i、L max,L和C。
针对上述情况一B、情况二B、情况三B和情况四，目标服务等级的配置参数包括队列资源参数及数据流约束参数；根据与目标服务等级关联的最大时延调整目标服务等级的参数，包括但不限于队列资源参数的第二更新值小于等于资源阈值的第一种情况下的调整方式，以及队列资源参数的第二更新值大于资源阈值的第二种情况下的调整方式，针对这两种情况的调整方式如下。
第一种情况下的调整方式：确定数据流约束参数的第二更新值；在最大时延不变的情况下，根据数据流约束参数的第二更新值确定队列资源参数的第二更新值；基于队列资源参数的第二更新值小于等于资源阈值，根据数据流约束参数的第二更新值调整之前数据流的准入约束参数，根据队列资源参数的第二更新值调整之前队列资源参数。
在示例性实施例中，队列资源参数的第二更新值包括队列带宽的第二更新值及队列缓存的第二更新值，数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值。
在最大时延不变的情况下，根据数据流约束参数的第二更新值确定队列资源参数的第二更新值，包括：在最大时延不变的情况下，确定目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量、第一网络设备的多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度；根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可 转发的第二最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第二更新值；根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，确定队列缓存的第二更新值。
示例性地，根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第二更新值，包括：根据目标服务等级对应的端口的带宽、目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量以及各个服务等级在目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定队列带宽的第二更新值
Figure PCTCN2020118590-appb-000083
Figure PCTCN2020118590-appb-000084
其中，C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000085
为目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量，Q j为第j服务等级可转发的最大数据量，
Figure PCTCN2020118590-appb-000086
为目标服务等级在目标服务等级的参数调整前可转发的最大数据量，
Figure PCTCN2020118590-appb-000087
为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，n为大于1的正整数。
示例性地，根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，确定队列缓存的第二更新值，包括：根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，按照如下公式确定队列缓存的第二更新值
Figure PCTCN2020118590-appb-000088
Figure PCTCN2020118590-appb-000089
其中，C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000090
为目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量，Q j为第j服务等级可转发的最大数据量，L max,L为低优先级队列中的最大报文长度，L n+1为非时延保障队列中的最大报文长度，L j为第j个服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000091
为数据流的突发量阈值的第二更新值，
Figure PCTCN2020118590-appb-000092
为目标服务等级的参数调整前数据流的突发量阈值，Δb i2为数据流的突发量阈值的增加值，
Figure PCTCN2020118590-appb-000093
为数据流的平均速率阈值的第二更新值，
Figure PCTCN2020118590-appb-000094
为目标服务等级的参数调整前数据流的平均速率阈值，Δr i2为数据流的平均速率阈值的增加值，L i为目标服务等级队列中的最大报文长度，n为大于1的正整数。
第二种情况下的调整方式一：目标服务等级的配置参数包括队列资源参数及数据流约束参数；根据目标服务等级关联的最大时延调整目标服务等级的参数，包括：确定目标服务等级的数据流约束参数的第二更新值；在最大时延不变的情况下，根据目标服务等级的数据流约束参数的第二更新值确定目标服务等级的队列资源参数的第二更新值；基于队列资源参数的第二更新值大于资源阈值，将目标服务等级对应的目标数据流切换至其他服务等级，其他服务等级包括第一网络设备的其他服务等级或者其他设备的服务等级。
示例性地，资源阈值包括但不限于非时延保障队列可抢占的资源大小，该资源包括但不限于带宽和缓存。
第二种情况下的调整方式二：目标服务等级的配置参数包括队列资源参数及数据流约束参数；根据目标服务等级关联的最大时延调整目标服务等级，包括：确定数据流约束参数的第二更新值；在最大时延不变的情况下，根据数据流约束参数的第二更新值确定队列资源参 数的第二更新值；基于队列资源参数的第二更新值大于资源阈值，根据数据流约束参数的第二更新值确定队列资源参数的第三更新值，确定可抢占资源的服务等级的参数的更新值，可抢占资源的服务等级的参数的更新值用于使得队列资源参数的第三更新值满足约束条件；根据数据流约束参数的第二更新值调整数据流约束参数，根据队列资源参数的第三更新值调整队列资源参数，根据可抢占资源的服务等级的参数的更新值调整可抢占资源的服务等级的参数。
示例性地，数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值；根据数据流约束参数的第二更新值确定队列资源参数的第三更新值，包括：确定目标服务等级对应的端口的带宽、目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量、第一网络设备的多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度；根据目标服务等级对应的端口的带宽、目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第三更新值；根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，确定队列缓存的第三更新值；
确定可抢占资源的服务等级的参数的更新值，包括：根据目标服务等级对应的端口的带宽、可抢占资源的服务等级调整后可转发的最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定可抢占资源的服务等级的队列带宽的更新值；根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、可抢占资源的服务等级的数据流的突发量阈值的更新值及数据流的平均速率阈值的更新值，确定可抢占资源的服务等级的队列缓存的更新值。
示例性地，根据目标服务等级对应的端口的带宽、目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定队列带宽的第三更新值，包括：根据目标服务等级对应的端口的带宽、目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定队列带宽的第三更新值
Figure PCTCN2020118590-appb-000095
Figure PCTCN2020118590-appb-000096
根据目标服务等级对应的端口的带宽、可抢占资源的服务等级调整后可转发的最大数据量以及多个服务等级在目标服务等级的参数调整前可转发的最大数据量，确定可抢占资源的服务等级的队列带宽
Figure PCTCN2020118590-appb-000097
Figure PCTCN2020118590-appb-000098
其中，C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000099
为目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量，
Figure PCTCN2020118590-appb-000100
为可抢占资源的服务等级调整后可转发的最大数据量，Q j为第j服务等级可转发的最大数据量，
Figure PCTCN2020118590-appb-000101
为目标服务等级的参数调整前可转发的最大数 据量，
Figure PCTCN2020118590-appb-000102
为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，n为大于1的正整数。
示例性地，根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，确定队列缓存的第三更新值，包括：
根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，按照如下公式确定队列缓存的第三更新值
Figure PCTCN2020118590-appb-000103
Figure PCTCN2020118590-appb-000104
根据多个服务等级在目标服务等级的参数调整前可转发的最大数据量以及多个服务等级对应的队列的最大报文长度、可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、可抢占资源的服务等级的数据流的突发量阈值的更新值及数据流的平均速率阈值的更新值，确定可抢占资源的服务等级的队列缓存
Figure PCTCN2020118590-appb-000105
Figure PCTCN2020118590-appb-000106
其中，C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000107
为目标服务等级在目标服务等级的参数调整后可转发的第三最大数据量，Q j为第j服务等级可转发的最大数据量，L max,L为低优先级队列中的最大报文长度，L n+1为非时延保障队列中的最大报文长度，L j为第j个服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000108
为数据流的突发量阈值的第二更新值，
Figure PCTCN2020118590-appb-000109
为目标服务等级的参数调整前数据流的突发量阈值，Δb i2为数据流的突发量阈值的增加值，
Figure PCTCN2020118590-appb-000110
为数据流的平均速率阈值的第二更新值，
Figure PCTCN2020118590-appb-000111
为目标服务等级的参数调整前数据流的平均速率阈值，Δr i2为数据流的平均速率阈值的增加值，L i为目标服务等级队列中的最大报文长度，L k为可抢占资源的服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000112
为可抢占资源的服务等级调整后可转发的最大数据量，
Figure PCTCN2020118590-appb-000113
为非时延保障队列在目标服务等级的参数调整后可转发的最大数据量，
Figure PCTCN2020118590-appb-000114
为可抢占资源的服务等级的数据流的平均速率阈值的更新值，
Figure PCTCN2020118590-appb-000115
为可抢占资源的服务等级的数据流的突发量阈值的更新值，n为大于1的正整数。
需要说明的是，上述第二种情况下的调整方式一以及调整方式二均是队列资源参数的第二更新值大于资源阈值的情况下的处理方式，在本申请实施例中，针对队列资源参数的第二更新值大于资源阈值的情况，采用调整方式一还是采用调整方式二，本申请实施例对此不进行限定。也就是说，队列资源参数的第二更新值大于资源阈值的情况下，既可以直接采用调整方式一，将目标服务等级对应的目标数据流切换至其他服务等级，也可以采用调整方式二，根据数据流约束参数的第二更新值调整数据流约束参数，根据队列资源参数的第三更新值调整队列资源参数，根据可抢占资源的服务等级的参数的更新值调整可抢占资源的服务等级的参数。
另外，在采用调整方式一仍然无法满足调整目标服务等级的参数所需资源的情况下，还可以继续采用调整方式二。示例性的，在采用调整方式二仍然无法满足调整目标服务等级的参数所需资源的情况下，还可以继续采用调整方式三。
无论是上述哪种情况，控制设备确定目标服务等级的参数的更新值之后，还包括：控制设备将调整的目标服务等级的参数的更新值发送给第一网络设备，指示第一网络设备按照目标服务等级的参数的更新值调整目标服务等级的参数。相应的，第一网络设备接收控制设备发送的目标服务等级的参数的更新值；按照目标服务等级的参数的更新值对目标服务等级的参数进行调整。
可选地，对于上述情况一B、情况二B、情况三B和情况四，第一网络设备确定目标服务等级参数的更新值后，若目标服务等级之外的其他服务等级参数也伴随确定了更新值，还包括：控制设备将调整目标服务等级之外的其他服务等级的参数的更新值发送给第一网络设备，指示第一网络设备按照目标服务等级之外的其他服务等级的参数的更新值调整对应的其他服务等级。相应的，第一网络设备接收控制设备发送的目标服务等级之外的其他服务等级的参数的更新值；按照其他服务等级的参数的更新值对对应的其他服务等级进行调整。
另外，需要说明的是，本申请实施例中的队列缓存的更新值是指所需的最小的队列缓存的值，在示例性实施例中，为了满足目标服务等级的调整要求，还可以采用大于队列缓存的更新值的其他值，本申请实施例对此不进行限定。
本申请实施例提供的方法，根据业务部署情况(例如当前数据流参数和剩余数据流参数)、队列状态信息及异常信息的上报次数等的变化，动态更新包括队列资源分配和数据流准入约束等服务等级的参数配置，以及对数据流经过的服务等级进行动态切换，在保障业务端到端时延的前提下，进一步提升网络资源利用效率。
基于上述图4所示的方法流程，以图3所示的系统架构为例，通过第一网络设备与控制器之间的交互过程，对本申请实施例提供的调整服务等级的方法进行说明。如图5所示，该方法包括如下501-511几个过程。
501，控制器全局业务部署单元111接收业务需求，为业务计算出路径和沿路需要进入的服务等级并进行部署。同时维护网络所有服务等级当前数据流参数和剩余允许准入数据流大小，包括对当前数据流参数的各项参数分别按照新业务签约承诺数据流的对应参数(数据流的突发量阈值、数据流的平均速率阈值等)进行增加，以及对当前剩余可允许准入流量的各项参数按照新业务签约承诺数据流的对应参数进行扣减。
502，各个网络设备的本地队列监控单元112周期性上报本地的各项队列状态信息给全局队列监控分析单元121，全局队列监控分析单元121维护全局所有服务等级的队列状态信息。
在示例性实施例中，第一网络设备的本地队列监控单元111周期性向控制器全局队列监控分析单元121上报本地的队列缓存占用、报文排队时延、或报文计数等队列状态信息。
503，网络设备本地异常信息上报单元125将本地QoS调整异常信息实时上报给控制器。
504，基于501、502、503，全局调整触发单元113根据全网服务等级当前数据流的准入情况、全局队列监控信息、本地QoS异常信息上报情况，触发全局目标服务等级的配置更新、或触发相关已部署数据流切换所经过的服务等级。
其中，触发全局目标服务等级的配置更新、或触发相关已部署数据流切换所经过的服务等级包括但不限于如下七种情形。
第一种情形：若根据队列监控信息，目标服务等级的队列缓存占用、报文排队时延、报文计数等任一队列状态信息在第一参考时间段内均小于对应的阈值下限，则全局调整触发单元113判断目标服务等级中的数据流的实际到达大小长期小于该目标服务等级签约承诺最大数据流大小，进而触发目标服务等级的队列资源、数据流准入等参数向减小方向进行配置更新。
第二种情形：若目标服务等级的当前数据流参数(数据流的突发量阈值和数据流的平均速率阈值)在第三参考时间段内均小于当前数据流参数对应的阈值下限，则全局调整触发单元113判断目标服务等级中的数据流部署长期处于不足状态，进而触发目标服务等级的队列资源、数据流准入等参数向减小方向进行配置更新。
第三种情形：若目标服务等级的剩余数据流参数(最大剩余允许突发和最大剩余平均速率)在第五参考时间段内均超过剩余数据流参数对应的阈值上限，则全局调整触发单元113判断目标服务等级中的数据流接近部署下限，进而触发目标服务等级的队列资源、数据流准入等参数向减小方向进行配置更新。
第四种情形：若根据队列监控信息，目标服务等级的队列缓存占用、报文排队时延、报文计数等任一队列状态信息在第二参考时间段内均大于对应的阈值上限，则全局调整触发单元113判断目标服务等级中实际到达的数据流超过该目标服务等级的数据流准入约束，进而触发目标服务等级的队列资源、数据流准入等参数向增大方向进行配置更新，或者触发经过目标服务等级的已部署数据流切换至该第一网络设备的其他服务等级或其他网络设备的服务等级。
第五种情形：若目标服务等级的当前数据流参数(数据流的突发量阈值或数据流的平均速率阈值)在第四参考时间段内均超过当前数据流参数对应的阈值上限，则全局调整触发单元113判断目标服务等级中的数据流部署接近部署上限，进而触发目标服务等级的队列资源、数据流准入等参数向增大方向进行配置更新，或触发经过目标服务等级的已部署数据流切换至该第一网络设备的其他服务等级或其他网络设备的服务等级。
第六种情形：若目标服务等级的剩余数据流参数(最大剩余允许突发或最大剩余平均速率)在第六参考时间段内均小于剩余数据流参数对应的阈值下限，则全局调整触发单元113判断目标服务等级中的数据流接近部署上限，进而触发目标服务等级的队列资源、数据流准入等参数向增大方向进行配置更新，或触发经过目标服务等级的已部署数据流切换至该第一网络设备的其他服务等级或其他网络设备的服务等级。
第七种情形：若接收到目标服务等级的数据流超准入约束ERROR信号上报次数超过目标服务等级的数据流超准入约束ERROR信号上报次数对应的阈值上限，则全局调整触发单元113判断目标服务等级实际到达的数据流超过该目标服务等级的数据流准入约束，且本地无法自行通过调整目标服务等级进行处理，进而触发目标服务等级的队列资源、数据流准入等参数向增大方向进行配置更新，或者触发经过目标服务等级的已部署数据流切换至该第一网络设备的其他服务等级或其他网络设备的服务等级。
505，基于504，全网调整计算单元114计算全网相关服务等级的更新配置，包括新的队列资源参数(队列带宽、队列缓存等)和数据流约束参数(数据流的突发量阈值、数据流的平均速率阈值等)；或者为全网相关已部署数据流计算新的沿路每跳需要经过的服务等级，在满足业务的端到端时延要求，以及每个服务等级的数据流准入约束的前提下，既可以选择将 原有某一服务等级切换到相同端口的另一服务等级，也可以选择将其切换到同一设备不同端口乃至不同设备的另一服务等级。
在示例性实施例中，全局调整计算单元114计算全网相关服务等级的更新配置包括但不限于如下两种方式。
第一种方式：
基于上述504中的第一种情形、第二种情形或第三种情形，全局调整计算单元114计算相关服务等级需要减少的数据流准入(包括需减少的数据流的突发量阈值、数据流的平均速率阈值等)，同时在保持关联的最大时延不变的前提下，计算目标服务等级相应需要减少的队列资源(包括需减少的队列带宽、队列缓存等)。
示例性地，全局调整计算单元114若基于网络演算理论计算服务等级的更新配置，原理如图6所示的出端口队列示意图，考虑在某出端口下共设置n个服务等级，每个服务等级各自绑定一个队列。另外考虑可以存在若干服务于无时延保障需求数据流的队列，为简化描述，用一个虚拟队列指代这些无时延保障需求队列，称之为非时延保障队列。所有服务等级对应的队列、以及非时延保障队列之间按照轮询方式调度。可选地，还可以存在若干低优先级队列，相对地，服务等级对应的队列和非时延保障队列则为高优先级队列。在每次调度机会下，高优先级队列都可进行优先调度。
以目标服务等级为第i个服务等级为例，在高优先级队列内部的轮询调度中，轮到第i个服务等级时，一次调度可以转发的最大数据量为Q i，轮到非时延保障队列时，一次调度可以转发的最大数据量为Q n+1。特别地，为方便统一表达，若当前端口下非时延保障队列不存在，可直接令Q n+1＝0。
如图7中的(1)所示，对于其中的第i个服务等级，定义其到达曲线函数α i(t)，自变量为t，在任意时刻t 1和t 2之间(t 2≥t 1≥0)的间隔t 2-t 1内，到达第i个服务等级的数据量不超过α i(t 2-t 1)。以第i个服务等级的队列从时刻0到时刻t累计到达的数据量为A i(t)为例，若存在函数α i(t)，对于任意时刻t 1和t 2(t 2≥t 1≥0)，都满足A i(t 2)-A i(t 1)≤α i(t 2-t 1)，则定义函数α i(t)为第i个服务等级的到达曲线。
此外，定义第i个服务等级的服务曲线函数β i(t)，自变量为t，在任意时刻t 1和t 2之间(t 2≥t 1≥0)的间隔t 2-t 1内，该服务等级成功转发的数据量不少于β i(t 2-t 1)。以第i个服务等级的队列从时刻0到时刻t累计发送的数据量为B i(t)，若存在函数β i(t)，对于任意时刻t 1和t 2(t 2≥t 1≥0)，都满足B i(t 2)-B i(t 1)≥β i(t 2-t 1)，则定义函数β i(t)为第i个服务等级的服务曲线。
示例性地，以目标服务等级为第i个服务等级为例，经过第i个服务等级的报文产生的最大时延可表示为D i＝sup t≥0{inf{d≥0|α i(t)≤β i(t+d)}}。以该端口总带宽为C，第i个服务等级的最大时延要求为D i，第i个服务等级对应队列中的最大报文长度为L i为例。此外，非时延保障队列中的最大报文长度为L n+1；低优先级队列中的最大报文长度为L max,L。若第i个服务等级需要进行配置更新，定义在配置更新之前每次轮到第i个服务等级时可以转发的最大数据量为
Figure PCTCN2020118590-appb-000116
非时延保障队列可以转发的最大数据量为
Figure PCTCN2020118590-appb-000117
特别地，若非时延保障队列在配置更新之前不存在，则令
Figure PCTCN2020118590-appb-000118
第i个服务等级配置更新之前，该第i个服务等级的出口带宽为
Figure PCTCN2020118590-appb-000119
相应地，在第i个服务等级配置更新之前，第i个服务等级的数据流的突发量阈值为
Figure PCTCN2020118590-appb-000120
平均速率为
Figure PCTCN2020118590-appb-000121
如图7中的(1)所示，
Figure PCTCN2020118590-appb-000122
表示第i个服务 等级更新前的到达曲线，此时第i个服务等级的服务曲线
Figure PCTCN2020118590-appb-000123
表示如下：
Figure PCTCN2020118590-appb-000124
其中，
Figure PCTCN2020118590-appb-000125
Figure PCTCN2020118590-appb-000126
是指取
Figure PCTCN2020118590-appb-000127
Figure PCTCN2020118590-appb-000128
的值与0的较大值。其中，T 0为时间偏移项，以最大时延D i为包含入端口排队时延、出端口排队时延、处理时延、发送时延的单个网络设备传输数据流的时延为例，则T 0定义为该网络设备的入端口排队时延、处理时延和发送时延之和；以最大时延D i为出端口排队时延为例，则T 0定义为0。
定义
Figure PCTCN2020118590-appb-000129
的横坐标截距为
Figure PCTCN2020118590-appb-000130
斜率为
Figure PCTCN2020118590-appb-000131
根据公式(1-1)，即有：
Figure PCTCN2020118590-appb-000132
Figure PCTCN2020118590-appb-000133
另一方面，第i个服务等级的最大时延值为D i表示如下：
Figure PCTCN2020118590-appb-000134
根据公式(1-2)、公式(1-3)和公式(1-4)，第i个服务等级的最大时延值为D i重新表示如下：
Figure PCTCN2020118590-appb-000135
相应地，在配置更新前，第i个服务等级为满足最大时延值D i要求所需的最小队列缓存大小
Figure PCTCN2020118590-appb-000136
表示如下：
Figure PCTCN2020118590-appb-000137
全局调整计算单元114计算出第i个服务等级需要减少的数据流的突发量阈值为Δb i，则更新后的数据流的突发量阈值的第一更新值为
Figure PCTCN2020118590-appb-000138
需要减少的数据流的平均速率阈值的减少值为Δr i，则更新后的数据流的平均速率阈值的第一更新值为
Figure PCTCN2020118590-appb-000139
在第i个服务等级配置更新之后，如图7中的(1)所示，
Figure PCTCN2020118590-appb-000140
表示第i个服务等级更新后的服务曲线，
Figure PCTCN2020118590-appb-000141
的横坐标截距为
Figure PCTCN2020118590-appb-000142
表示第i个服务等级更新后的到达曲线，此时保持第i个服务等级的最大时延值为D i不变，可重新表示如下：
Figure PCTCN2020118590-appb-000143
其中
Figure PCTCN2020118590-appb-000144
为新计算的轮到调度第i个服务等级时可转发的第一最大数据量，也即是目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量；
Figure PCTCN2020118590-appb-000145
为新计算的轮到非时延保障队列时可转发的最大数据量，也即是非时延保障队列在目标服务等级的参数调整后可转发的最大数据量。为了确保在第i个服务等级配置更新后，其他服务等级的时延要求不被影响，需要保证如下约束：
Figure PCTCN2020118590-appb-000146
根据公式(1-5)、公式(1-7)和公式(1-8)，可以计算出配置更新后轮到调度第i个服务等级时可转发的第一最大数据量
Figure PCTCN2020118590-appb-000147
如下：
Figure PCTCN2020118590-appb-000148
根据公式(1-8)和公式(1-9)，更新后非时延保障队列每次调度的最大数据量
Figure PCTCN2020118590-appb-000149
表示如下：
Figure PCTCN2020118590-appb-000150
根据公式(1-8)和公式(1-9)，计算出更新后的第i个服务等级的队列预留带宽，即队列带宽的第一更新值
Figure PCTCN2020118590-appb-000151
Figure PCTCN2020118590-appb-000152
Figure PCTCN2020118590-appb-000153
为目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量，Q j为第j服务等级可转发的最大数据量，
Figure PCTCN2020118590-appb-000154
为目标服务等级在目标服务等级的参数调整前可转发的最大数据量，
Figure PCTCN2020118590-appb-000155
为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，n为大于1的正整数。
根据公式(1-6)、公式(1-8)和公式(1-9)，配置更新后，第i个服务等级的最小所需队列缓存大小，即队列缓存的第一更新值
Figure PCTCN2020118590-appb-000156
计算如下：
Figure PCTCN2020118590-appb-000157
其中，L max,L为低优先级队列中的最大报文长度，C为目标服务等级对应的端口的带宽，
Figure PCTCN2020118590-appb-000158
为目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量，Q j为第j服务等级可转发的最大数据量，L n+1为非时延保障队列中的最大报文长度，L j为第j个服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000159
为数据流的突发量阈值的第一更新值，
Figure PCTCN2020118590-appb-000160
为目标服务等级的参数调整前数据流的突发量阈值，Δb i1为目标服务等级的数据流的突发量阈值的减少值，
Figure PCTCN2020118590-appb-000161
为数据流的平均速率阈值的第一更新值，
Figure PCTCN2020118590-appb-000162
为目标服务等级的参数调整前数据流的平均速率阈值，Δr i1为数据流的平均速率阈值的减少值，L i为目标服务等级队列中的最大报文长度，
Figure PCTCN2020118590-appb-000163
为非时延保障队列在目标服务等级的参数调整后可转发的最大数据量，n为大于1的正整数。
在示例性实施例中，若第一网络设备同端口下非时延保障队列当前可抢占的带宽、缓存等资源足够满足目标服务等级额外需要增加的队列资源需求，则本地配置更新执行单元124抢占非时延保障队列的相应资源，将其转移给目标服务等级，按照新的队列带宽和缓存要求对目标服务等级的参数进行更新。
第二种方式：
基于上述504中的第四种情形、第五种情形、第六种情形或第七种情形，全局调整计算单元114计算相关服务等级额外需要增加的数据流准入(包括需增加的数据流的突发量阈值、数据流的平均速率阈值等)，同时在保持目标服务等级关联的最大时延不变的前提下，计算目标服务等级相应需要增加的队列资源(包括需增加的队列预留带宽、缓存等)。计算结果需要保证：目标服务等级增加的队列资源，不超过同端口下可抢占的非时延保障队列的资源(带宽、缓存)和同端口下其他服务等级在不影响自身最大时延要求前提下可释放的队列资源的总和。
示例性地，全局调整计算单元114若基于网络演算理论计算服务等级的更新配置。原理如图6的端口示意图所示。相关参数可参见上述第一种方式中的描述，此处不再赘述。
示例性地，第i个服务等级配置更新之前，该第i个服务等级的出口带宽为
Figure PCTCN2020118590-appb-000164
Figure PCTCN2020118590-appb-000165
相应地，在第i个服务等级配置更新之前，第i个服务等级的数据流的突发量阈值为
Figure PCTCN2020118590-appb-000166
平均速率为
Figure PCTCN2020118590-appb-000167
如图7中的(2)所示，
Figure PCTCN2020118590-appb-000168
表示第i个服务等级更新前的到达曲线，此时第i个服务等级的服务曲线
Figure PCTCN2020118590-appb-000169
表示如下：
Figure PCTCN2020118590-appb-000170
其中，
Figure PCTCN2020118590-appb-000171
Figure PCTCN2020118590-appb-000172
是指取
Figure PCTCN2020118590-appb-000173
Figure PCTCN2020118590-appb-000174
的值与0的较大值。其中，T 0为时间偏移项，以最大时延D i为包含入端口排队时延、出端 口排队时延、处理时延、发送时延的单个网络设备传输数据流的时延为例，则T 0定义为该网络设备的入端口排队时延、处理时延和发送时延之和；以最大时延D i为出端口排队时延为例，则T 0定义为0。
第i个服务等级的最大时延值为D i表示如下：
Figure PCTCN2020118590-appb-000175
相应地，在配置更新前，第i个服务等级为满足最大时延值D i要求所需的最小队列缓存大小
Figure PCTCN2020118590-appb-000176
表示如下：
Figure PCTCN2020118590-appb-000177
全局调整计算单元114计算出第i个服务等级需要增加的数据流的突发量阈值为Δb i2，则更新后的数据流的突发量阈值的第二更新值为
Figure PCTCN2020118590-appb-000178
需要增加的数据流的平均速率阈值为Δr i2，则更新后的数据流的平均速率阈值的第二更新值为
Figure PCTCN2020118590-appb-000179
示例性地，对于目标服务等级需要增加的队列资源，优先抢占同端口下可抢占的非时延保障队列的资源(带宽、缓存)，若同端口下可抢占的非时延保障队列的资源不满足要求，进一步抢占同端口下其他服务等级在不影响自身最大时延要求前提下可释放的队列资源。具体而言，1)若同端口下非时延保障队列存在：1a)若非时延保障队列可抢占资源足够满足目标服务等级调整要求，目标服务等级直接抢占可满足调整要求的最少量的非时延保障队列资源；1b)若非时延保障队列可抢占资源不足以满足目标服务等级调整要求，目标服务等级首先抢占非时延保障队列所有可抢占的资源，进而依次抢占同端口下其他服务等级在不影响自身最大时延要求前提下可释放的队列资源，直到满足目标服务等级的调整要求。2)若同端口下非时延保障队列不存在，直接依次抢占同端口下其他服务等级在不影响自身最大时延要求前提下可释放的队列资源，直到满足目标服务等级的调整要求。
情形一：存在非时延保障队列。首先判断非时延保障队列的可抢占资源是否满足目标服务等级的调整要求。在保持第i个服务等级的最大时延值D i不变的前提下，假设为了满足更新的流量准入(数据流的突发量阈值的第二更新值
Figure PCTCN2020118590-appb-000180
数据流的平均速率阈值的第二更新值
Figure PCTCN2020118590-appb-000181
)，轮到调度第i个服务等级时可转发的最大数据量需要更新为
Figure PCTCN2020118590-appb-000182
相应地，轮到调度非时延保障队列时可转发的最大数据量需要更新为
Figure PCTCN2020118590-appb-000183
例如，在第i个服务等级配置更新之后，如图7中的(2)所示，
Figure PCTCN2020118590-appb-000184
表示第i个服务等级更新后的服务曲线，
Figure PCTCN2020118590-appb-000185
的横坐标截距为
Figure PCTCN2020118590-appb-000186
表示第i个服务等级更新后的到达曲线，此时D i可重新表示如下：
Figure PCTCN2020118590-appb-000187
为了确保在第i个服务等级调整后，其他服务等级的时延要求不被影响，需要保证如下约束：
Figure PCTCN2020118590-appb-000188
根据公式(2-2)、公式(2-4)和公式(2-5)，可以求得假设满足调整要求时需更新的轮到调度第i个服务等级时可转发的最大数据量
Figure PCTCN2020118590-appb-000189
如下：
Figure PCTCN2020118590-appb-000190
根据公式(2-5)和公式(2-6)，假设满足调整要求时需更新的非时延保障队列每次调度的最大数据量
Figure PCTCN2020118590-appb-000191
表示如下：
Figure PCTCN2020118590-appb-000192
根据公式(2-5)和公式(2-6)，计算出假设满足调整要求时第i个服务等级的队列预留带宽
Figure PCTCN2020118590-appb-000193
需更新为：
Figure PCTCN2020118590-appb-000194
根据公式(2-3)、公式(2-5)和公式(2-6)，计算出假设满足调整要求时第i个服务等级的最小所需队列缓存大小
Figure PCTCN2020118590-appb-000195
需更新为：
Figure PCTCN2020118590-appb-000196
情形一A：定义非时延保障队列当前最大可抢占的带宽资源为ΔC n+1，最大可抢占的缓存资源为ΔBF n+1。若非时延保障队列当前可抢占资源满足目标服务等级的调整要求，即满足如下两个条件：
Figure PCTCN2020118590-appb-000197
Figure PCTCN2020118590-appb-000198
此时可确定目标服务等级的队列带宽的第二更新值
Figure PCTCN2020118590-appb-000199
同时确定队列缓存的第二更新值
Figure PCTCN2020118590-appb-000200
其中，队列带宽的第二更新值
Figure PCTCN2020118590-appb-000201
和队列缓存的第二更新值
Figure PCTCN2020118590-appb-000202
的确定过程也可参见402中第一种情况下的调整方式，队列带宽的第二更新值
Figure PCTCN2020118590-appb-000203
Figure PCTCN2020118590-appb-000204
其中，
Figure PCTCN2020118590-appb-000205
的确定过程可参考上述
Figure PCTCN2020118590-appb-000206
的确定过程。
队列缓存的第二更新值
Figure PCTCN2020118590-appb-000207
Figure PCTCN2020118590-appb-000208
其中，
Figure PCTCN2020118590-appb-000209
的确定过程可参考上述
Figure PCTCN2020118590-appb-000210
的确定过程。
情形一B：若非时延保障队列当前可抢占资源不满足目标服务等级的调整要求，根据目标服务等级关联的最大时延调整目标服务等级，包括：确定数据流约束参数的第二更新值；在最大时延不变的情况下，根据数据流约束参数的第二更新值确定队列资源参数的第二更新值；基于队列资源参数的第二更新值大于资源阈值，将目标服务等级对应的目标数据流切换至其他服务等级，其他服务等级包括第一网络设备的其他服务等级或者其他设备的服务等级。
或者，目标服务等级首先抢占非时延保障队列所有可抢占的资源，进而依次抢占同端口下其他服务等级在不影响自身最大时延要求前提下可释放的队列资源，直到满足目标服务等级的调整要求。
示例性地，为简化分析，假设非时延保障队列当前可抢占的缓存资源足够大，始终能满足目标服务等级的调整要求，以下主要描述带宽资源的调整过程。考虑非时延保障队列当前可抢占的带宽资源不足，无法满足目标服务等级的调整要求，即有：
Figure PCTCN2020118590-appb-000211
其中
Figure PCTCN2020118590-appb-000212
和
Figure PCTCN2020118590-appb-000213
为上述计算出的假设满足调整要求前提下目标服务等级和非时延保障队列需要调整成的带宽资源。首先将非时延保障队列的带宽资源调整为允许调整的最大值
Figure PCTCN2020118590-appb-000214
并作为非时延保障队列最终确定的带宽更新值。此时轮到调度非时延保障队列时可转发的最大数据量的更新值确定为：
Figure PCTCN2020118590-appb-000215
根据公式(2-5)，目标服务等级的带宽资源需调整为
Figure PCTCN2020118590-appb-000216
相应地，此时轮到调度目标服务等级时可转发的最大数据量需调整为：
Figure PCTCN2020118590-appb-000217
由于抢占非时延保障队列的带宽资源无法满足目标服务等级的调整要求，目标服务等级此时的调整参数
Figure PCTCN2020118590-appb-000218
和
Figure PCTCN2020118590-appb-000219
仅为临时状态，需要进一步考虑抢占同端口下其他服务等级在不影响自身最大时延要求前提下可释放的队列资源。为方便描述，将抢占非时延保障队列或者抢占某一个个同端口下的其他队列的相应队列资源，都看成一次迭代过程。则抢占非时延保障队列相应队列资源视为第一次迭代过程，为统一表示，重新将
Figure PCTCN2020118590-appb-000220
表示为
Figure PCTCN2020118590-appb-000221
根据公式(2-6)，在第一次迭代过程下目标服务等级准入流量的数据流的突发量阈值
Figure PCTCN2020118590-appb-000222
需满足：
Figure PCTCN2020118590-appb-000223
考虑选择同端口下其他服务等级中的任意一个进行抢占，开始第二次迭代过程。假设选中的为第k个服务等级，在第二次迭代过程中，除了目标服务等级和第k个服务等级之外，保持当前其他服务等级的参数Q j和非时延保障队列的参数
Figure PCTCN2020118590-appb-000224
不变。定义第k个服务等级的更新前的总流量准入参数为
Figure PCTCN2020118590-appb-000225
和
Figure PCTCN2020118590-appb-000226
分别为更新前的数据流的突发量阈值和数据流的平均速率阈值，同时定义第k个服务等级当前允许准入流量参数为
Figure PCTCN2020118590-appb-000227
和
Figure PCTCN2020118590-appb-000228
分别为当前允许准入突发量阈值和当前允许准入平均速率阈值，根据定义有
Figure PCTCN2020118590-appb-000229
需要指出的是，对其他服务等级的选择方式在本申请实施例中不进行限定，可基于经验或者某些规则，如优先选择相对更为空闲的服务等级等。
在第二次迭代过程中，假设满足调整要求时需更新的轮到调度第i个服务等级时可转发的最大数据量定义为
Figure PCTCN2020118590-appb-000230
对应的，假设满足调整要求时需更新的轮到调度第k个服务等级时可转发的最大数据量定义为
Figure PCTCN2020118590-appb-000231
类比于公式(2-5)，为了保证第i个服务等级和第k个服务等级调整时，其他服务等级的最大时延不受影响，需要满足一下约束：
Figure PCTCN2020118590-appb-000232
其中
Figure PCTCN2020118590-appb-000233
即为更新前轮到调度第k个服务等级时可转发的最大数据量。
类比于公式(2-6)，在第二次迭代过程中，假设满足调整要求时需更新的轮到调度第i个服务等级时可转发的最大数据量
Figure PCTCN2020118590-appb-000234
表示如下：
Figure PCTCN2020118590-appb-000235
类比于公式(2-7)，在第二次迭代过程中，假设满足调整要求时需更新的轮到调度第k个服务等级时可转发的最大数据量
Figure PCTCN2020118590-appb-000236
表示如下：
Figure PCTCN2020118590-appb-000237
如果第k个服务等级在不影响自身最大时延要求前提下可释放的队列资源足够满足目标服务等级的调整要求，也即是说按照
Figure PCTCN2020118590-appb-000238
和
Figure PCTCN2020118590-appb-000239
调整目标服务等级和第k个服务等级后，第k个服务等级的当前允许准入流量的最大时延仍然满足要求，即满足如下条件：
Figure PCTCN2020118590-appb-000240
Figure PCTCN2020118590-appb-000241
此时可最终确定轮到调度目标服务等级时可转发的最大数据量
Figure PCTCN2020118590-appb-000242
最终确定目标服务等级的队列带宽的第三更新值
Figure PCTCN2020118590-appb-000243
为：
Figure PCTCN2020118590-appb-000244
同时最终确定轮到调度第k个服务等级时可转发的最大数据量
Figure PCTCN2020118590-appb-000245
最终确定第k个服务等级的队列带宽的更新值
Figure PCTCN2020118590-appb-000246
为：
Figure PCTCN2020118590-appb-000247
同时最终确定目标服务等级的队列缓存的第三更新值
Figure PCTCN2020118590-appb-000248
为：
Figure PCTCN2020118590-appb-000249
同时最终确定第k个服务等级的队列缓存的更新值
Figure PCTCN2020118590-appb-000250
为：
Figure PCTCN2020118590-appb-000251
其中
Figure PCTCN2020118590-appb-000252
和
Figure PCTCN2020118590-appb-000253
是第k个服务等级的流量准入参数的更新值，可由如下公式计算出具体值：
Figure PCTCN2020118590-appb-000254
Figure PCTCN2020118590-appb-000255
如果第k个服务等级在不影响自身最大时延要求前提下可释放的队列资源不足以满足目标服务等级的调整要求，也即是说按照
Figure PCTCN2020118590-appb-000256
和
Figure PCTCN2020118590-appb-000257
调整目标服务等级和第k个服务等级后，第k个服务等级的当前允许准入流量的最大时延无法满足要求，即满足如下条件：
Figure PCTCN2020118590-appb-000258
此时首先将轮到调度第k个服务等级时可转发的最大数据量调整为允许调整的最大值
Figure PCTCN2020118590-appb-000259
并将其作为第k个服务等级最终确定的带宽更新值。类比公式(2-15)，
Figure PCTCN2020118590-appb-000260
计算为：
Figure PCTCN2020118590-appb-000261
Figure PCTCN2020118590-appb-000262
类比于公式(2-16)，为了保证第i个服务等级和第k个服务等级调整时，其他服务等级的最大时延不受影响，需要满足一下约束：
Figure PCTCN2020118590-appb-000263
则第二次迭代过程下，轮到调度目标服务等级的最大转发数据量更新为
Figure PCTCN2020118590-appb-000264
在第二次迭代过程下，目标服务等级准入流量的数据流的突发量阈值
Figure PCTCN2020118590-appb-000265
需满足：
Figure PCTCN2020118590-appb-000266
由于经过第二次迭代过程，抢占非时延保障队列和第k个服务等级可释放的资源之后，仍然无法满足目标服务等级的调整要求，继续考虑选择同端口下除目标服务等级和第k个服务等级之外其他服务等级中的任意一个进行抢占，开始第三次迭代过程。具体过程和第二次迭代过程原理相同，此处不再赘述具体细节。该迭代过程可反复进行，直到某一次的迭代过程中，服务等级的调整要求得到满足，服务等级调整过程结束。不失一般性，假设最终除了目标服务等级和非时延保障队列之外，还有K个其他服务等级j 1～j K也进行了调整，具体而言，按照第j 1个、第j 2个……第j K个服务等级的顺序进行迭代调整。假设最终确定轮到调度目标服务等级时可转发的最大数据量为
Figure PCTCN2020118590-appb-000267
轮到非时延保障队列时可转发的最大数据量为
Figure PCTCN2020118590-appb-000268
轮到其他第j k个服务等级时可转发的最大数据量为
Figure PCTCN2020118590-appb-000269
其中1≤k≤K，最终确定目标服务等级的队列带宽的第三更新值
Figure PCTCN2020118590-appb-000270
为：
Figure PCTCN2020118590-appb-000271
同时最终确定非时延保障队列的队列带宽的更新值
Figure PCTCN2020118590-appb-000272
为：
Figure PCTCN2020118590-appb-000273
同时最终确定任意第j k个服务等级的队列带宽的更新值
Figure PCTCN2020118590-appb-000274
为：
Figure PCTCN2020118590-appb-000275
同时最终确定目标服务等级的队列缓存的第三更新值
Figure PCTCN2020118590-appb-000276
为：
Figure PCTCN2020118590-appb-000277
同时最终确定任意第j k个服务等级的队列缓存的更新值
Figure PCTCN2020118590-appb-000278
为：
Figure PCTCN2020118590-appb-000279
Figure PCTCN2020118590-appb-000280
其中，
Figure PCTCN2020118590-appb-000281
是第j k个服务等级的数据流的平均速率阈值的更新值，
Figure PCTCN2020118590-appb-000282
是第j k个服务等级的数据流的突发量阈值的更新值，
Figure PCTCN2020118590-appb-000283
Figure PCTCN2020118590-appb-000284
其中
Figure PCTCN2020118590-appb-000285
是量化符号，
Figure PCTCN2020118590-appb-000286
是第j k个服务等级当前允许准入的平均速率阈值，
Figure PCTCN2020118590-appb-000287
是第j k个服务等级当前允许准入的突发量阈值。
对于第j K个服务等级的流量准入参数
Figure PCTCN2020118590-appb-000288
和
Figure PCTCN2020118590-appb-000289
可由如下公式计算出具体值：
Figure PCTCN2020118590-appb-000290
Figure PCTCN2020118590-appb-000291
需要说明的是，以上仅以一次迭代和两次迭代为例进行说明，在示例性实施例中，可能还有第三次迭代等更多次的迭代，原理同上述第二次迭代一致，本申请实施例不再一一赘述。
情形二：非时延保障队列不存在。直接依次抢占同端口下其他服务等级在不影响自身最大时延要求前提下可释放的队列资源，直到满足目标服务等级的调整要求。
示例性地，可以直接令轮到调度非时延保障队列时最大可转发数据量
Figure PCTCN2020118590-appb-000292
基本计算过程和情形一相同，仅仅是省去了抢占非时延保障队列资源的第一次迭代过程，直接从抢占同端口下其他服务等级的资源的第二次迭代过程开始，此处不再赘述。
在示例性实施例中，已部署数据流经过某些数据流接近部署上限或数据流超发的目标服务等级，全局调整计算单元114针对这些数据流计算新的沿路每跳要经过的服务等级，同时保证按该结果部署仍然满足业务的端到端时延需求和各个服务等级的数据流准入约束。对于需要切换的目标服务等级，优先考虑切换至本地设备同端口其他相对空闲的服务等级。若无可行计算结果，进一步考虑切换至本地设备其他端口乃至其他网络设备相对空闲的其他服务等级。
在示例性实施例中，全局调整计算单元114令数据流服务等级切换先于服务等级更新配置执行。若计算出目标服务等级额外需要增加的队列资源(包括队列带宽、缓存等)超过同端口下可抢占的非时延保障队列的资源(带宽、缓存)和同端口下其他服务等级可释放队列资源的总和，则针对经过其他服务等级的相关已部署数据流进一步计算需要切换的服务等级，使得其他服务等级可释放队列资源增加，从而满足相关服务等级的额外队列资源需求。规定在执行上述服务等级配置更新之前，预先执行上述已部署数据流的目标服务等级切换。
在示例性实施例中，全局调整计算单元114令目标服务等级更新配置先于数据流服务等级切换执行。若计算出相关已部署数据流的路径切换结果，但待切换至的服务等级的剩余数据流准入不足，则针对该服务等级进一步计算更新配置，使该目标服务等级能够相关数据流的切换需求。规定在执行上述已部署数据流的服务等级切换之前，预先执行上述目标服务等级的配置更新。
506，基于505，控制器将全网调整计算单元114计算出的目标服务等级更新配置结果下发 至相关设备的本地配置更新执行单元124。特别地，若目标服务等级之外的其他服务等级的参数也伴随计算出更新值，控制器也将对应的其他服务等级的参数的更新值下发至相关设备的本地配置更新执行单元124。服务等级更新配置信息包括新的队列资源参数(队列预留带宽、队列缓存等)、数据流约束参数(数据流的突发量阈值、数据流的平均速率阈值等)。
507，基于506，本地配置更新执行单元124对目标服务等级实施相应的配置更新。
在示例性实施例中，本地配置更新执行单元124基于上述505中的第一种方式的计算结果实施目标服务等级的配置更新，优先考虑将同端口下可抢占的非时延保障队列的带宽、缓存资源转移给目标服务等级。若仍不满足目标服务等级额外需要增加的队列资源需求，继续考虑将同端口下其他相对空闲的服务等级在不影响自身最大时延要求前提下可减少的队列资源转移给目标服务等级，直至实现指定要求的配置更新。
在示例性实施例中，本地配置更新执行单元124基于上述505中的第二种方式的计算结果实施目标服务等级的配置更新，将目标服务等级释放出的队列资源转移为同端口下可抢占的非时延保障队列的带宽、缓存资源。
508，基于507，全局服务等级维护单元116对目标服务等级的参数的更新值进行同步。后续控制器基于目标服务等级的参数的更新值进行业务部署。
509，基于505，全局服务等级切换单元115按照全网调整计算单元114计算出的服务等级切换结果对相关数据流需要经过的服务等级实施切换，后续相关数据流的报文按照新的服务等级进行传输。
510，基于508和509，全局业务部署单元111按需对相关业务的签约承诺数据流参数按照实际到达情况进行修正，包括修改数据流的突发量阈值、数据流的平均速率阈值等，同时结合相关服务等级的更新配置或相关业务的服务等级切换情况，按需修改各服务等级当前允许准入的数据流大小和剩余可允许准入的数据流大小。
在示例性实施例中，全局业务部署单元111对目标服务等级内超出签约承诺的数据流的参数按照实际到达情况进行向上修正，包括向上修正数据流的突发量阈值、数据流的平均速率阈值等；对目标服务等级内长期例如目标参考时间段内均小于签约承诺的数据流的参数按照实际到达情况进行向下修正，包括向下修正数据流的突发量阈值、数据流的平均速率阈值等。
本申请实施例提供了一种调整服务等级的装置，该装置通过图8所示的模块实现图4和图5所示的调整服务等级的方法。参见图8，该装置包括：
获取模块801用于获取第一网络设备的目标服务等级的相关信息，相关信息包括目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及异常信息的上报次数中的至少一种信息。示例性的，该获取模块801所执行的功能可参见图4中的401所述的相关内容，或者参考图5所示的501-503所述的相关内容，此处不再一一赘述。
调整模块802用于基于相关信息中的任一信息不满足与任一信息对应的阈值，根据与目标服务等级关联的最大时延调整目标服务等级的参数。
示例性的，该调整模块802所执行的功能可参见图4中的402所述的相关内容，或者参考图5所示的504-510所述的相关内容，此处不再一一赘述。
本申请实施例提供的方法，根据业务部署情况(例如当前数据流参数和剩余数据流参数)、队列状态信息及异常信息的上报次数等的变化，动态更新包括队列资源分配和数据流准入约束等服务等级的参数配置，以及对数据流经过的服务等级进行动态切换，在保障业务端到端 时延的前提下，进一步提升网络资源利用效率。
在一些实施例中，上述图8提供的装置在实现其功能时，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将设备的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。另外，上述实施例提供的装置与方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
本申请实施例提出了一种调整服务等级的系统，能够在保证业务端到端时延需求的前提下，进一步提升网络资源利用率。在进行QoS动态调整时，本申请实施例能够保证网络中的数据流不会产生时延劣化或丢包。如图2所示，一种基于服务等级、保障端到端时延的QoS调整系统，在保障业务端到端时延的前提下，提升网络资源利用效率。该系统的整体架构包含控制器侧和网络设备侧的若干单元，以及各个单元的相关功能及相互间的交互逻辑。
整体工作流程，包含控制器全局触发QoS调整流程，及相应的原始信息收集、调整触发判断、调整计算、调整执行、调整信息同步等步骤，可以根据业务部署情况、数据流实际到达情况、队列状态信息等的变化，实现对服务等级参数的动态配置更新、以及对数据流经过的服务等级的动态切换。在进行服务等级的动态配置更新、或数据流的服务等级切换时，不会引起网络数据流的时延劣化或丢包。
服务等级参数的配置更新包括以下一个或多个参数：增加或减小队列资源分配，如队列预留带宽、缓存等；增加或减小数据流约束参数，如数据流的突发量阈值、汇聚流最大平均速率等。数据流经过服务等级的切换包括：切换至同设备同端口下其他服务等级；切换至同设备其他端口或其他设备的服务等级。
在一种示例中，网络设备向控制设备发送目标服务等级的相关信息，其中，相关信息包括所述目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及错误信息的上报次数中的至少一种信息。控制设备获取第一网络设备的目标服务等级的相关信息。控制设备基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。控制设备所执行的功能可参见图4中的402所述的相关内容，或者参考图5所示的504-510所述的相关内容，此处不再一一赘述。
上述实施例中控制设备或者网络设备，可以为路由器或交换机。硬件结构包括但不限于如下两种：
1、如图9所示，控制设备或网络设备包括收发器、处理器和存储器。
示例性的，针对控制设备，该控制设备的收发器用于接收报文或数据信息等，例如，参见图4所示的401或者图5的501-503所述的相关内容，该控制设备的收发器用于接收第一网络设备上报的第一网络设备的目标服务等级的至少一个队列状态信息、当前数据流参数以及异常信息的上报次数等信息，还可接收处理器中的存储的一些参数，例如剩余数据流参数。则该控制设备的处理器用于基于相关信息中的任一信息不满足与该任一信息对应的阈值，根据与目标服务等级关联的最大时延调整目标服务等级的参数，例如，参见图4所示的402的相关描述，或者图5所示的504-510所述的控制设备调整目标服务等级的参数的相关描述。
2、如图10所示，控制设备或网络设备包括主控板和接口板，主控板包括处理器和存储器，接口板包括处理器、存储器和接口卡。接口板的处理器用于调用接口板的存储器中的程序指 令执行报文的接收和发送。主控板的处理器用于调用主控板存储器中的程序指令执行相应的处理功能。
示例性的，针对控制设备，该控制设备的接口板的接口卡用于接收报文或数据信息等，例如，参见图4所示的401或者图5的501-503所述的相关内容，该控制设备的接口板的接口卡用于接收第一网络设备上报的第一网络设备的目标服务等级的至少一个队列状态信息、当前数据流参数以及异常信息的上报次数等信息，还可接收接口板的处理器中存储的一些参数，例如剩余数据流参数。则该控制设备的主控板的处理器用于基于相关信息中的任一信息不满足与该任一信息对应的阈值，根据与目标服务等级关联的最大时延调整目标服务等级的参数，例如，参见图4所示的402的相关描述，或者图5所示的504-510所述的控制设备调整目标服务等级的参数的相关描述。
以上描述的任意装置实施例都仅仅是示意性的，其中作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。另外，本发明提供的第一网络节点或控制器实施例附图中，模块之间的连接关系表示它们之间具有通信连接，具体可以实现为一条或多条通信总线或信号线。本领域普通技术人员在不付出创造性劳动的情况下，即可以理解并实施。
本申请实施例公开内容所描述的方法或者算法的步骤可以硬件的方式来实现，也可以是由处理器执行软件指令的方式来实现。软件指令可以由相应的软件模块组成，软件模块可以被存放于随机存取存储器(random access memory，RAM)、闪存、只读存储器(read only memory，ROM)、可擦除可编程只读存储器(erasable programmable ROM，EPROM)、电可擦可编程只读存储器(electrically EPROM，EEPROM)、硬盘、移动硬盘、光盘或者本领域熟知的任何其它形式的存储介质中。一种示例性的存储介质耦合至处理器，从而使处理器能够从该存储介质读取信息，且可向该存储介质写入信息。当然，存储介质也可以是处理器的组成部分。处理器和存储介质可以位于ASIC中。另外，该ASIC可以位于核心网接口设备中。当然，处理器和存储介质也可以作为分立组件存在于核心网接口设备中。
在上述一个或多个示例中，本申请所描述的功能可以用硬件、软件、固件或它们的任意组合来实现。当使用软件实现时，可以将这些功能存储在计算机可读介质中或者作为计算机可读介质上的一个或多个指令或代码进行传输。计算机可读介质包括计算机存储介质和通信介质，其中通信介质包括便于从一个地方向另一个地方传送计算机程序的任何介质。存储介质可以是通用或专用计算机能够存取的任何可用介质。
图11为本申请实施例的调整服务等级的设备1100的硬件结构示意图。图11所示的调整服务等级的设备1100可以执行上述图3所示实施例提供的配置数据的管理方法中的相应步骤。
如图11所示，调整服务等级的设备1100包括处理器1101、存储器1102、接口1103和总线1104。其中接口1103可以通过无线或有线的方式实现，示例性地，该接口1103可以是网卡。上述处理器1101、存储器1102和接口1103通过总线1104连接。
接口1103可以包括发送器和接收器，用于与其他通信设备通信，执行图4所示的401或图5所示的相关步骤。处理器1101用于执行上述图4所示实施例中402或图5所示的处理相关步骤。处理器1101和/或用于本文所描述的技术的其他过程。存储器1102包括操作系统11021和应用程序11022，用于存储程序、代码或指令，当处理器或硬件设备执行这些程序、 代码或指令时可以完成方法实施例中涉及调整服务等级的设备1100的处理过程。可选的，存储器1102可以包括只读存储器(英文：Read-only Memory，缩写：ROM)和随机存取存储器(英文：Random Access Memory，缩写：RAM)。其中，ROM包括基本输入/输出系统(英文：Basic Input/Output System，缩写：BIOS)或嵌入式系统；RAM包括应用程序和操作系统。当需要运行调整服务等级的设备1100时，通过固化在ROM中的BIOS或者嵌入式系统中的bootloader引导系统进行启动，引导调整服务等级的设备1100进入正常运行状态。在调整服务等级的设备1100进入正常运行状态后，运行在RAM中的应用程序和操作系统，从而，完成方法实施例中涉及调整服务等级的设备1100的处理过程。
可以理解的是，图11仅仅示出了调整服务等级的设备1100的简化设计。在实际应用中，调整服务等级的设备1100可以包含任意数量的接口，处理器或者存储器。
应理解的是，上述处理器可以是中央处理器(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(digital signal processing，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现场可编程门阵列(field－programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者是任何常规的处理器等。值得说明的是，处理器可以是支持进阶精简指令集机器(advanced RISC machines，ARM)架构的处理器。
进一步地，在一种可选的实施例中，上述存储器可以包括只读存储器和随机存取存储器，并向处理器提供指令和数据。存储器还可以包括非易失性随机存取存储器。例如，存储器还可以存储设备类型的信息。
该存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用。例如，静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic random access memory，DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data date SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。
本申请还提供了一种计算机可读存储介质，存储介质中存储有至少一条指令，指令由处理器加载并执行以实现如上任一所述的调整服务等级的方法。例如，可以执行图4或图5中的方法。
本申请提供了一种计算机程序，当计算机程序被计算机执行时，可以使得处理器或计算机可以执行图4或图5中的方法实施例中对应的各个步骤和/或流程。
提供了一种芯片，包括处理器，用于从存储器中调用并运行所述存储器中存储的指令，使得安装有所述芯片的通信设备可以执行图4或图5中的方法。
提供另一种芯片，包括：输入接口、输出接口、处理器和存储器，所述输入接口、输出 接口、所述处理器以及所述存储器之间通过内部连接通路相连，所述处理器用于执行所述存储器中的代码，当所述代码被执行时，所述处理器用于执行上述各方面中的方法。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线)或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如固态硬盘Solid State Disk)等。
以上所述的具体实施方式，对本申请的目的、技术方案和有益效果进行了进一步详细说明，所应理解的是，以上所述仅为本申请的具体实施方式而已，并不用于限定本申请的保护范围，凡在本申请的技术方案的基础之上，所做的任何修改、等同替换、改进等，均应包括在本申请的保护范围之内。

Claims (52)

  1. 一种调整服务等级的方法，其特征在于，所述方法包括：
    控制设备获取第一网络设备的目标服务等级的相关信息，所述相关信息包括所述目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及异常信息的上报次数中的至少一种信息；
    基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  2. 根据权利要求1所述的方法，其特征在于，所述相关信息包括所述目标服务等级的至少一个队列状态信息；
    所述基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数，包括：
    基于所述至少一个队列状态信息中的任一队列状态信息小于与所述任一队列状态信息对应的阈值下限，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  3. 根据权利要求1所述的方法，其特征在于，所述相关信息包括所述目标服务等级的当前数据流参数；
    所述基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数，包括：
    基于所述目标服务等级的当前数据流参数小于与所述当前数据流参数对应的阈值下限，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  4. 根据权利要求1所述的方法，其特征在于，所述相关信息包括所述目标服务等级的剩余数据流参数；
    所述基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数，包括：
    基于所述目标服务等级的剩余数据流参数超过与所述剩余数据流参数对应的阈值上限，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  5. 根据权利要求2-4任一所述的方法，其特征在于，所述目标服务等级的参数包括队列资源参数及数据流约束参数；
    所述根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数，包括：
    确定所述数据流约束参数的第一更新值；
    在所述最大时延不变的情况下，根据所述数据流约束参数的第一更新值确定所述队列资源参数的第一更新值；
    根据所述数据流约束参数的第一更新值调整所述数据流约束参数，根据所述队列资源参数的第一更新值调整所述队列资源参数。
  6. 根据权利要求5所述的方法，其特征在于，所述队列资源参数的第一更新值包括队列带宽的第一更新值及队列缓存的第一更新值，所述数据流约束参数的第一更新值包括数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值；
    在所述最大时延不变的情况下，根据所述数据流约束参数的第一更新值确定所述队列资源参数的第一更新值，包括：
    在所述最大时延不变的情况下，确定所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第一最大数据量、所述第一网络设备的多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度；
    根据所述端口的带宽、所述第一最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第一更新值；
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述最大报文长度、所述数据流的突发量阈值的第一更新值及所述数据流的平均速率阈值的第一更新值，确定所述队列缓存的第一更新值。
  7. 根据权利要求6所述的方法，其特征在于，所述根据所述目标服务等级对应的端口的带宽、所述第一最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第一更新值，包括：
    根据所述目标服务等级对应的端口的带宽、所述第一最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定所述队列带宽的第一更新值
    Figure PCTCN2020118590-appb-100001
    Figure PCTCN2020118590-appb-100002
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100003
    为所述第一最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100004
    为所述目标服务等级在所述目标服务等级的参数调整前可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100005
    为非时延保障队列在所述目标服务等级的参数调整前可转发的最大数据量，所述n、i和j为正整数。
  8. 根据权利要求6或7所述的方法，其特征在于，所述根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述最大报文长度、所述数据流的突发量阈值的第一更新值及所述数据流的平均速率阈值的第一更新值，确定所述队列缓存的第一更新值，包括：
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述最大报文长度、所述数据流的突发量阈值的第一更新值及所述数据流的平均速率阈值的第一更新值，按照如下公式确定所述队列缓存的第一更新值
    Figure PCTCN2020118590-appb-100006
    Figure PCTCN2020118590-appb-100007
    其中，所述L max,L为低优先级队列中的最大报文长度，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100008
    为所述第一最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述L n+1为非时延保障队列中的最大报文长度，所述L j为第j个服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100009
    为所述数据流的突发量阈值的第一更新值，所述
    Figure PCTCN2020118590-appb-100010
    所述
    Figure PCTCN2020118590-appb-100011
    为所述目标服务等级的参数调整前数据流的突发量阈值，所述Δb i1为所述目标服务等级的数据流的突发量阈值的减少值，所述
    Figure PCTCN2020118590-appb-100012
    为所述数据流的平均速率阈值的第一更新值，所述
    Figure PCTCN2020118590-appb-100013
    所述
    Figure PCTCN2020118590-appb-100014
    为所述目标服务等级的参数调整前数据流的平均速率阈值，所述Δr i1为数据流的平均速率阈值的减少值，所述L i为所述目标服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100015
    为非时延保障队 列在所述目标服务等级的参数调整后可转发的最大数据量，所述n、i和j为正整数。
  9. 根据权利要求1所述的方法，其特征在于，所述相关信息包括所述目标服务等级的至少一个队列状态信息；
    所述基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数，包括：
    基于所述至少一个队列状态信息中的任一队列状态信息超过与所述任一队列状态信息对应的阈值上限，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  10. 根据权利要求1所述的方法，其特征在于，所述相关信息包括所述目标服务等级的当前数据流参数；
    所述基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数，包括：
    基于所述目标服务等级的当前数据流参数超过与所述当前数据流参数对应的阈值上限，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  11. 根据权利要求1所述的方法，其特征在于，所述相关信息包括所述目标服务等级的剩余数据流参数；
    所述基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数，包括：
    基于所述目标服务等级的剩余数据流参数小于与所述剩余数据流参数对应的阈值下限，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  12. 根据权利要求1所述的方法，其特征在于，所述相关信息包括所述目标服务等级的异常信息的上报次数；
    所述基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数，包括：
    基于所述目标服务等级的异常信息的上报次数超过与所述异常信息的上报次数对应的阈值上限，所述控制设备根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  13. 根据权利要求9-12任一所述的方法，其特征在于，所述目标服务等级的参数包括队列资源参数及数据流约束参数；
    所述根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数，包括：
    确定所述数据流约束参数的第二更新值；
    在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值；
    基于所述队列资源参数的第二更新值小于或等于资源阈值，根据所述数据流约束参数的第二更新值调整所述数据流约束参数，根据所述队列资源参数的第二更新值调整所述队列资源参数。
  14. 根据权利要求9-12任一所述的方法，其特征在于，所述目标服务等级的配置参数包括队列资源参数及数据流约束参数；
    所述根据与所述目标服务等级关联的最大时延调整所述目标服务等级，包括：
    确定所述数据流约束参数的第二更新值；
    在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值；
    基于所述队列资源参数的第二更新值大于资源阈值，将所述目标服务等级对应的目标数据流切换至其他服务等级，所述其他服务等级包括所述第一网络设备的其他服务等级或者其他设备的服务等级。
  15. 根据权利要求9-12任一所述的方法，其特征在于，所述目标服务等级的配置参数包括队列资源参数及数据流约束参数；
    所述根据与所述目标服务等级关联的最大时延调整所述目标服务等级，包括：
    确定所述数据流约束参数的第二更新值；在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值；
    基于所述队列资源参数的第二更新值大于资源阈值，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第三更新值，确定可抢占资源的服务等级的参数的更新值，所述可抢占资源的服务等级的参数的更新值用于使得所述队列资源参数的第三更新值满足约束条件；
    根据所述数据流约束参数的第二更新值调整所述数据流约束参数，根据所述队列资源参数的第三更新值调整所述队列资源参数，根据所述可抢占资源的服务等级的参数的更新值调整所述可抢占资源的服务等级的参数。
  16. 根据权利要求13-15任一所述的方法，其特征在于，所述队列资源参数的第二更新值包括队列带宽的第二更新值及队列缓存的第二更新值，所述数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值；所述在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值，包括：
    在所述最大时延不变的情况下，确定所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第二最大数据量、所述第一网络设备的多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度；
    根据所述端口的带宽、所述第二最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第二更新值；
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，确定所述队列缓存的第二更新值。
  17. 根据权利要求16所述的方法，其特征在于，所述根据所述目标服务等级对应的端口的带宽、所述第二最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第二更新值，包括：
    根据所述目标服务等级对应的端口的带宽、所述第二最大数据量以及所述各个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定所述队列带宽的第二更新值
    Figure PCTCN2020118590-appb-100016
    Figure PCTCN2020118590-appb-100017
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100018
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第二最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100019
    为所述目标服务等级在所述目标服务等级的参数调整前可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100020
    为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，所述n、i和j为正整数。
  18. 根据权利要求16或17所述的方法，其特征在于，所述根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，确定所述队列缓存的第二更新值，包括：
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，按照如下公式确定所述队列缓存的第二更新值
    Figure PCTCN2020118590-appb-100021
    Figure PCTCN2020118590-appb-100022
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100023
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第二最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述L max,L为低优先级队列中的最大报文长度，所述L n+1为非时延保障队列中的最大报文长度，所述L j为第j个服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100024
    为所述数据流的突发量阈值的第二更新值，所述
    Figure PCTCN2020118590-appb-100025
    所述
    Figure PCTCN2020118590-appb-100026
    为所述目标服务等级的参数调整前数据流的突发量阈值，所述Δb i2为所述数据流的突发量阈值的增加值，所述
    Figure PCTCN2020118590-appb-100027
    为所述数据流的平均速率阈值的第二更新值，所述
    Figure PCTCN2020118590-appb-100028
    所述
    Figure PCTCN2020118590-appb-100029
    为所述目标服务等级的参数调整前数据流的平均速率阈值，所述Δr i2为所述数据流的平均速率阈值的增加值，所述L i为目标服务等级队列中的最大报文长度，所述n、i和j为正整数。
  19. 根据权利要求15所述的方法，其特征在于，所述数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值；
    所述根据所述数据流约束参数的第二更新值确定所述队列资源参数的第三更新值，包括：
    确定所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量、所述第一网络设备的多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度；
    根据所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第三更新值；
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，确定所述队列缓存的第三更新值；
    所述确定可抢占资源的服务等级的参数的更新值，包括：
    根据所述目标服务等级对应的端口的带宽、所述可抢占资源的服务等级调整后可转发的最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述可抢占资源的服务等级的队列带宽的更新值；
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述可抢占资源的服务等级的数据流的突发量阈值的更新值及数据流的平均速率阈值的更新值，确定所述可抢占资源的服务等级的队列缓存的更新值。
  20. 根据权利要求19所述的方法，其特征在于，所述根据所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第三更新值，包括：
    根据所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定所述队列带宽的第三更新值
    Figure PCTCN2020118590-appb-100030
    Figure PCTCN2020118590-appb-100031
    根据所述目标服务等级对应的端口的带宽、所述可抢占资源的服务等级调整后可转发的最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述可抢占资源的服务等级的队列带宽
    Figure PCTCN2020118590-appb-100032
    Figure PCTCN2020118590-appb-100033
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100034
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量，所述
    Figure PCTCN2020118590-appb-100035
    为所述可抢占资源的服务等级调整后可转发的最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100036
    为所述目标服务等级在所述目标服务等级的参数调整前可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100037
    为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，所述n、i和j为正整数。
  21. 根据权利要求19或20所述的方法，其特征在于，所述根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，确定所述队列缓存的第三更新值，包括：
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，按照如下公式确定所述队列缓存的第三更新值
    Figure PCTCN2020118590-appb-100038
    Figure PCTCN2020118590-appb-100039
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述可抢占资源的服务等级的数据流的突发量阈值的更新值及数据流的平均速率阈值的更新值，确定所述可抢占资源的服务等级的队列缓存
    Figure PCTCN2020118590-appb-100040
    Figure PCTCN2020118590-appb-100041
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100042
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述L max,L为低优先级队列中的最大报文长度，所述L n+1为非时延保障队列中的最大报文长度，所述L j为第j个服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100043
    为所述数据流的突发量阈值的第二更新值，所述
    Figure PCTCN2020118590-appb-100044
    所述
    Figure PCTCN2020118590-appb-100045
    为所述目标服务等级的参数调整前数据流的突发量阈值，所述Δb i2为所述数据流的突发量阈值的增加值，所述
    Figure PCTCN2020118590-appb-100046
    为所述数据流的平均速率阈值的第二更新值，所述
    Figure PCTCN2020118590-appb-100047
    所述
    Figure PCTCN2020118590-appb-100048
    为所述目标服务等级的参数调整前数据流的平均速率阈值，所述Δr i2为所述数据流的平均速率阈值的增加值，所述L i为目标服务等级队列中的最大报文长度，所述L k为所述可抢占资源的服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100049
    为所述可抢占资源的服务等级调整后可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100050
    为非时延保障队列在目标服务等级的参数调整后可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100051
    为所述可抢占资源的服务等级的数据流的平均速率阈值的更新值，所述
    Figure PCTCN2020118590-appb-100052
    为所述可抢占资源的服务等级的数据流的突发量阈值的更新值，所述n、i和j为正整数。
  22. 根据权利要求1-21任一所述的方法，其特征在于，所述第一网络设备调整所述目标服务等级的参数之后，还包括：
    所述控制设备将调整的所述目标服务等级的参数的更新值发送给所述第一网络设备，指示所述第一网络设备按照所述目标服务等级的参数的更新值调整所述目标服务等级的参数。
  23. 根据权利要求1-22任一所述的方法，其特征在于，所述控制设备获取第一网络设备的目标服务等级的相关信息包括：
    所述控制设备在预设周期内接收所述第一网络设备发送的所述目标服务等级的相关信息。
  24. 根据权利要求1-23中任一所述的方法，其特征在于，所述队列状态信息包括队列的缓存占用、报文排队时延和队列的报文计数中的一个或多个。
  25. 一种调整服务等级的装置，其特征在于，所述装置包括：
    获取模块，用于获取第一网络设备的目标服务等级的相关信息，所述相关信息包括所述目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及异常信息的上报次数中的至少一种信息；
    调整模块，用于基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  26. 根据权利要求25所述的装置，其特征在于，所述相关信息包括所述目标服务等级的至少一个队列状态信息；
    所述调整模块，用于基于所述至少一个队列状态信息中的任一队列状态信息小于与所述任一队列状态信息对应的阈值下限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  27. 根据权利要求26所述的装置，其特征在于，所述相关信息包括所述目标服务等级的当前数据流参数；
    所述调整模块，用于基于所述目标服务等级的当前数据流参数小于与所述当前数据流参数对应的阈值下限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  28. 根据权利要求25所述的装置，其特征在于，所述相关信息包括所述目标服务等级的剩余数据流参数；
    所述调整模块，用于基于所述目标服务等级的剩余数据流参数超过与所述剩余数据流参数对应的阈值上限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  29. 根据权利要求26-28任一所述的装置，其特征在于，所述目标服务等级的参数包括队列资源参数及数据流约束参数；
    所述调整模块，用于确定所述数据流约束参数的第一更新值；在所述最大时延不变的情况下，根据所述数据流约束参数的第一更新值确定所述队列资源参数的第一更新值；根据所述数据流约束参数的第一更新值调整所述数据流约束参数，根据所述队列资源参数的第一更新值调整所述队列资源参数。
  30. 根据权利要求29所述的装置，其特征在于，所述队列资源参数的第一更新值包括队列带宽的第一更新值及队列缓存的第一更新值，所述数据流约束参数的第一更新值包括数据流的突发量阈值的第一更新值及数据流的平均速率阈值的第一更新值；
    所述调整模块，用于在所述最大时延不变的情况下，确定所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第一最大数据量、所述第一网络设备的多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度；
    根据所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第一更新值；
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第一更新值及所述数据流的平均速率阈值的第一更新值，确定所述队列缓存的第一更新值。
  31. 根据权利要求30所述的装置，其特征在于，所述调整模块，用于根据所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第一最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定所述队列带宽的第一更新值
    Figure PCTCN2020118590-appb-100053
    Figure PCTCN2020118590-appb-100054
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100055
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第一最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100056
    为所述目标服务等级在所述目标服务等级的参数调整前可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100057
    为非时延保障队列在所述目标服务等级的参数调整前可转发的最大数据量，所述n、i和j为正整数。
  32. 根据权利要求30或31所述的装置，其特征在于，所述调整模块，用于根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第一更新值及所述数据流的平均速率阈值的第一更新值，按照如下公式确定所述队列缓存的第一更新值
    Figure PCTCN2020118590-appb-100058
    Figure PCTCN2020118590-appb-100059
    其中，所述L max,L为低优先级队列中的最大报文长度，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100060
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第一最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述L n+1为非时延保障队列中的最大报文长度，所述L j为第j个服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100061
    为所述数据流的突发量阈值的第一更新值，所述
    Figure PCTCN2020118590-appb-100062
    所述
    Figure PCTCN2020118590-appb-100063
    为所述目标服务等级的参数调整前数据流的突发量阈值，所述Δb i1为所述目标服务等级的数据流的突发量阈值的减少值，所述
    Figure PCTCN2020118590-appb-100064
    为数据流的平均速率阈值的第一更新值，所述
    Figure PCTCN2020118590-appb-100065
    所述
    Figure PCTCN2020118590-appb-100066
    为所述目标服务等级的参数调整前数据流的平均速率阈值，所述Δr i1为数据流的平均速率阈值的减少值，所述L i为所述目标服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100067
    为非时延保障队列在所述目标服务等级的参数调整后可转发的最大数据量，所述n、i和j为正整数。
  33. 根据权利要求25所述的装置，其特征在于，所述相关信息包括所述目标服务等级的至少一个队列状态信息；
    所述调整模块，用于基于所述至少一个队列状态信息中的任一队列状态信息超过与所述任一队列状态信息对应的阈值上限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  34. 根据权利要求25所述的装置，其特征在于，所述相关信息包括所述目标服务等级的当前数据流参数；
    所述调整模块，用于基于所述目标服务等级的当前数据流参数超过与所述当前数据流参数对应的阈值上限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  35. 根据权利要求25所述的装置，其特征在于，所述相关信息包括所述目标服务等级的剩余数据流参数；
    所述调整模块，用于基于所述目标服务等级的剩余数据流参数小于与所述剩余数据流参数对应的阈值下限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  36. 根据权利要求25所述的装置，其特征在于，所述相关信息包括所述目标服务等级的异常信息的上报次数；
    所述调整模块，用于基于所述目标服务等级的异常信息的上报次数超过与所述异常信息的上报次数对应的阈值上限，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  37. 根据权利要求33-36任一所述的装置，其特征在于，所述目标服务等级的参数包括队列资源参数及数据流约束参数；
    所述调整模块，用于确定所述数据流约束参数的第二更新值；
    在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值；
    基于所述队列资源参数的第二更新值小于等于资源阈值，根据所述数据流约束参数的第二更新值调整所述数据流约束参数，根据所述队列资源参数的第二更新值调整所述队列资源参数。
  38. 根据权利要求33-36任一所述的装置，其特征在于，所述目标服务等级的配置参数包括队列资源参数及数据流约束参数；
    所述调整模块，用于确定所述数据流约束参数的第二更新值；
    在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值；
    基于所述队列资源参数的第二更新值大于资源阈值，将所述目标服务等级对应的目标数据流切换至其他服务等级，所述其他服务等级包括所述第一网络设备的其他服务等级或者其他设备的服务等级。
  39. 根据权利要求33-36任一所述的装置，其特征在于，所述目标服务等级的配置参数包括队列资源参数及数据流约束参数；
    所述调整模块，用于确定所述数据流约束参数的第二更新值；在所述最大时延不变的情况下，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第二更新值；
    基于所述队列资源参数的第二更新值大于资源阈值，根据所述数据流约束参数的第二更新值确定所述队列资源参数的第三更新值，确定可抢占资源的服务等级的参数的更新值，所述可抢占资源的服务等级的参数的更新值用于使得所述队列资源参数的第三更新值满足约束条件；
    根据所述数据流约束参数的第二更新值调整所述数据流约束参数，根据所述队列资源参数的第三更新值调整所述队列资源参数，根据所述可抢占资源的服务等级的参数的更新值调整所述可抢占资源的服务等级的参数。
  40. 根据权利要求37-39任一所述的装置，其特征在于，所述队列资源参数的第二更新值包括队列带宽的第二更新值及队列缓存的第二更新值，所述数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值；
    所述调整模块，用于在所述最大时延不变的情况下，确定所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量、所述第一网络设备的多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度；
    根据所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第二更新值；
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，确定所述队列缓存的第二更新值。
  41. 根据权利要求40所述的装置，其特征在于，所述调整模块，用于根据所述目标服务等级对应的端口的带宽、所述目标服务等级在目标服务等级的参数调整后可转发的第二最大数据量以及所述各个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定所述队列带宽的第二更新值
    Figure PCTCN2020118590-appb-100068
    Figure PCTCN2020118590-appb-100069
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100070
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第二最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100071
    为所述目标服务等级在所述目标服务等级的参数调整前可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100072
    为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，所述n、i和j为正整数。
  42. 根据权利要求40或41所述的装置，其特征在于，所述调整模块，用于根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值，按照如下公式确定所述队列缓存的第二更新值
    Figure PCTCN2020118590-appb-100073
    Figure PCTCN2020118590-appb-100074
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100075
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第二最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述L max,L为低优先级队列中的最大报文长度，所述L n+1为非时延保障队列中的最大报文长度，所述L j为第j个服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100076
    为所述数据流的突发量阈值的第二更新值，所述
    Figure PCTCN2020118590-appb-100077
    所述
    Figure PCTCN2020118590-appb-100078
    为所述目标服务等级的参数调整前数据流的突发量阈值，所述Δb i2为所述数据流的突发量阈值的增加值，所述
    Figure PCTCN2020118590-appb-100079
    为所述数据流的平均速率阈值的第二更新值，所述
    Figure PCTCN2020118590-appb-100080
    所述
    Figure PCTCN2020118590-appb-100081
    为所述目标服务等级的参数调整前数据流的平均速率阈值，所述Δr i2为所述数据流的平均速率阈值的增加值，所述L i为目标服务等级队列中的最大报文长度，所述n、i和j为正整数。
  43. 根据权利要求39所述的装置，其特征在于，所述数据流约束参数的第二更新值包括数据流的突发量阈值的第二更新值及数据流的平均速率阈值的第二更新值；
    所述调整模块，用于确定所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量、所述第一网络设备的多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度；
    根据所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述队列带宽的第三更新值；
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，确定所述队列缓存的第三更新值；
    所述调整模块，用于根据所述目标服务等级对应的端口的带宽、所述可抢占资源的服务等级调整后可转发的最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述可抢占资源的服务等级的队列带宽的更新值；
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述可抢占资源的服务等级的数据流的突发量阈值的更新值及数据流的平均速率阈值的更新值，确定所述可抢占资源的服务等级的队列缓存的更新值。
  44. 根据权利要求43所述的装置，其特征在于，所述调整模块，用于根据所述目标服务等级对应的端口的带宽、所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，按照如下公式确定所述队列带宽的第三更新值
    Figure PCTCN2020118590-appb-100082
    Figure PCTCN2020118590-appb-100083
    根据所述目标服务等级对应的端口的带宽、所述可抢占资源的服务等级调整后可转发的最大数据量以及所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量，确定所述可抢占资源的服务等级的队列带宽
    Figure PCTCN2020118590-appb-100084
    Figure PCTCN2020118590-appb-100085
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100086
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量，所述
    Figure PCTCN2020118590-appb-100087
    为所述可抢占资源的服务等级调整后可转发的最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100088
    为所述目标服务等级在所述目标服务等级的参数调整前可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100089
    为非时延保障队列在目标服务等级的参数调整前可转发的最大数据量，所述n、i和j为正整数。
  45. 根据权利要求43或44所述的装置，其特征在于，所述调整模块，用于根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述数据流的突发量阈值的第二更新值及所述数据流的平均速率阈值的第二更新值，按照如下公式确定所述队列缓存的第三更新值
    Figure PCTCN2020118590-appb-100090
    Figure PCTCN2020118590-appb-100091
    根据所述多个服务等级在所述目标服务等级的参数调整前可转发的最大数据量以及所述多个服务等级对应的队列的最大报文长度、所述可抢占资源的服务等级调整后可转发的最大数据量及最大报文长度、所述可抢占资源的服务等级的数据流的突发量阈值的更新值及数据流的平均速率阈值的更新值，确定所述可抢占资源的服务等级的队列缓存
    Figure PCTCN2020118590-appb-100092
    Figure PCTCN2020118590-appb-100093
    其中，所述C为所述目标服务等级对应的端口的带宽，所述
    Figure PCTCN2020118590-appb-100094
    为所述目标服务等级在所述目标服务等级的参数调整后可转发的第三最大数据量，所述Q j为第j服务等级可转发的最大数据量，所述L max,L为低优先级队列中的最大报文长度，所述L n+1为非时延保障队列中的最大报文长度，所述L j为第j个服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100095
    为所述数据流的突发量阈值的第二更新值，所述
    Figure PCTCN2020118590-appb-100096
    所述
    Figure PCTCN2020118590-appb-100097
    为所述目标服务等级的参数调整前数据流的突发量阈值，所述Δb i2为所述数据流的突发量阈值的增加值，所述
    Figure PCTCN2020118590-appb-100098
    为所述数据流的平均速率阈值的第二更新值，所述
    Figure PCTCN2020118590-appb-100099
    所述
    Figure PCTCN2020118590-appb-100100
    为所述目标服务等级的参数调整前数据流的平均速率阈值，所述Δr i2为所述数据流的平均速率阈值的增加值，所述L i为目标服务等级队列中的最大报文长度，所述L k为所述可抢占资源的服务等级队列中的最大报文长度，所述
    Figure PCTCN2020118590-appb-100101
    为所述可抢占资源的服务等级调整后可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100102
    为非时延保障队列在目标 服务等级的参数调整后可转发的最大数据量，所述
    Figure PCTCN2020118590-appb-100103
    为所述可抢占资源的服务等级的数据流的平均速率阈值的更新值，所述
    Figure PCTCN2020118590-appb-100104
    为所述可抢占资源的服务等级的数据流的突发量阈值的更新值，所述n、i和j为正整数。
  46. 根据权利要求25-45任一所述的装置，其特征在于，所述装置，还包括：
    发送模块，用于将调整的所述目标服务等级的参数的更新值发送给所述第一网络设备，指示所述第一网络设备按照所述目标服务等级的参数的更新值调整所述目标服务等级的参数。
  47. 根据权利要求25-46任一所述的装置，其特征在于，所述获取模块，用于在预设周期内接收所述第一网络设备发送的目标服务等级的相关信息。
  48. 根据权利要求25-47中任一所述的装置，其特征在于，所述队列状态信息包括队列的缓存占用、报文排队时延和队列的报文计数中的一个或多个。
  49. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质包括程序，当所述程序在计算机上执行时，使得所述计算机执行如权利要求1-24任一所述的方法。
  50. 一种网络设备，其特征在于，所述网络设备包括存储器及处理器，所述存储器中存储有至少一条指令，所述至少一条指令由所述处理器加载并执行，以实现如权利要求1-24任一所述的方法。
  51. 一种调整服务等级的系统，其特征在于，所述系统包括控制设备和第一网络设备；
    所述第一网络设备，用于发送第一网络设备的目标服务等级的相关信息；
    所述控制设备，用于接收所述目标服务等级的相关信息，所述目标服务等级的相关信息包括所述目标服务等级的至少一个队列状态信息、剩余数据流参数、当前数据流参数以及异常信息的上报次数中的至少一种信息；
    所述控制设备，还用于基于所述相关信息中的任一信息不满足与所述任一信息对应的阈值，根据与所述目标服务等级关联的最大时延调整所述目标服务等级的参数。
  52. 根据权利要求51所述的系统，其特征在于，所述系统包括权利要求25-48任一所述的控制设备。
PCT/CN2020/118590 2020-01-22 2020-09-28 调整服务等级的方法、装置、设备、系统及存储介质 WO2021147368A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
KR1020227028085A KR102693696B1 (ko) 2020-01-22 2020-09-28 서비스 레벨 조절 방법, 장치, 디바이스, 및 시스템, 그리고 저장 매체
JP2022544401A JP7531594B2 (ja) 2020-01-22 2020-09-28 サービスレベル調整方法、装置、デバイスおよびシステム、ならびに記憶媒体
EP20915290.9A EP4084409A4 (en) 2020-01-22 2020-09-28 Method and apparatus for adjusting service level, device, system and storage medium
US17/814,071 US20220407808A1 (en) 2020-01-22 2022-07-21 Service Level Adjustment Method, Apparatus, Device, and System, and Storage Medium

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN202010075494.5 2020-01-22
CN202010075494 2020-01-22
CN202010635155.8A CN113162789A (zh) 2020-01-22 2020-07-03 调整服务等级的方法、装置、设备、系统及存储介质
CN202010635155.8 2020-07-03

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/814,071 Continuation US20220407808A1 (en) 2020-01-22 2022-07-21 Service Level Adjustment Method, Apparatus, Device, and System, and Storage Medium

Publications (1)

Publication Number Publication Date
WO2021147368A1 true WO2021147368A1 (zh) 2021-07-29

Family

ID=76882225

Family Applications (2)

Application Number Title Priority Date Filing Date
PCT/CN2020/118590 WO2021147368A1 (zh) 2020-01-22 2020-09-28 调整服务等级的方法、装置、设备、系统及存储介质
PCT/CN2021/073420 WO2021148020A1 (zh) 2020-01-22 2021-01-22 调整服务等级的方法、装置、设备及存储介质

Family Applications After (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/073420 WO2021148020A1 (zh) 2020-01-22 2021-01-22 调整服务等级的方法、装置、设备及存储介质

Country Status (6)

Country Link
US (2) US20220407808A1 (zh)
EP (2) EP4084409A4 (zh)
JP (2) JP7531594B2 (zh)
KR (2) KR102693696B1 (zh)
CN (2) CN113162790B (zh)
WO (2) WO2021147368A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112787952B (zh) * 2019-11-08 2024-07-02 华为技术有限公司 一种业务流量调整方法及装置
CN113162790B (zh) * 2020-01-22 2023-10-03 华为技术有限公司 调整服务等级的方法、装置、设备及存储介质
US20230327950A1 (en) * 2020-09-09 2023-10-12 Lenovo (Singapore) Pte. Limited Method and Apparatus Including Event Triggered Goal Change for Control Loops
CN114070776B (zh) * 2021-11-02 2023-05-02 中国联合网络通信集团有限公司 一种改进的时间敏感网络数据传输方法、装置及设备
CN114301817A (zh) * 2021-12-17 2022-04-08 中电信数智科技有限公司 基于Netconf协议的设备监测阈值设置方法和系统
CN114726598B (zh) * 2022-03-28 2024-02-09 新华三信息安全技术有限公司 一种影响因子调整方法及装置
CN115996195B (zh) * 2023-03-23 2023-05-30 腾讯科技（深圳）有限公司 一种数据传输方法、装置、设备及介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1918937A (zh) * 2004-02-03 2007-02-21 诺基亚公司 提供端对端服务质量的方法和设备
US20080247357A1 (en) * 2005-11-18 2008-10-09 Sk Telecom Co., Ltd. Method For Adaptive Delay Threshold-Based Priority Queueing Scheme For Packet Scheduling in Mobile Broadband Wireless Access System
CN109245936A (zh) * 2012-09-06 2019-01-18 华为技术有限公司 控制网络传输时延的方法、服务质量控制实体和通信设备

Family Cites Families (85)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6567416B1 (en) * 1997-10-14 2003-05-20 Lucent Technologies Inc. Method for access control in a multiple access system for communications networks
AU3883499A (en) * 1998-05-07 1999-11-23 Aprisma Management Technologies, Inc. Multiple priority buffering in a computer network
US6990529B2 (en) * 2000-02-24 2006-01-24 Zarlink Semiconductor V.N., Inc. Unified algorithm for frame scheduling and buffer management in differentiated services networks
US6990113B1 (en) * 2000-09-08 2006-01-24 Mitsubishi Electric Research Labs., Inc. Adaptive-weighted packet scheduler for supporting premium service in a communications network
JP2002344509A (ja) * 2001-05-17 2002-11-29 Nippon Telegr & Teleph Corp <Ntt> ルータとパケットの読み出しレート制御方法およびその処理プログラム
JP2003110605A (ja) * 2001-09-28 2003-04-11 Mitsubishi Electric Corp ポリシー制御システム、ポリシー制御方法およびその方法をコンピュータに実行させるプログラム
US7263065B1 (en) * 2002-08-30 2007-08-28 At&T Corp. Network having multiple QoS levels
US7436765B2 (en) * 2003-09-03 2008-10-14 Teknovus, Inc. Method and apparatus for dynamically allocating upstream bandwidth in passive optical networks
US7362704B2 (en) * 2003-09-15 2008-04-22 Teknovus, Inc. Method and apparatus for dynamically allocating upstream bandwidth in passive optical networks
FI115100B (fi) * 2003-10-14 2005-02-28 Tellabs Oy Menetelmä ja laitteisto ruuhkanhallinnan sekä siirtoyhteyskapasiteetin vuorottamisen ohjaamiseksi pakettikytkentäisessä tietoliikenteessä
US7394780B1 (en) * 2003-11-26 2008-07-01 Idirect Incorporated Method, apparatus, and system for downstream recovery in a communication network
US7680139B1 (en) * 2004-03-25 2010-03-16 Verizon Patent And Licensing Inc. Systems and methods for queue management in packet-switched networks
US7720063B2 (en) * 2004-07-02 2010-05-18 Vt Idirect, Inc. Method apparatus and system for accelerated communication
DE602006019833D1 (de) * 2005-03-14 2011-03-10 Koninkl Philips Electronics Nv Messung und überwachung der dienstqualität in dienstdifferenzierten drahtlosen netzwerken
US7836250B2 (en) * 2005-07-15 2010-11-16 International Business Machines Corporation Automatic rebalancing of a data storage system
US8036372B2 (en) * 2005-11-30 2011-10-11 Avaya Inc. Methods and apparatus for dynamically reallocating a preferred request to one or more generic queues
US7779142B1 (en) * 2007-01-23 2010-08-17 Juniper Networks, Inc. Bandwidth allocation to support fast buffering
CN101309165B (zh) * 2007-05-14 2012-04-04 华为技术有限公司 信息报告的控制方法、装置和设备
US8688129B2 (en) * 2007-09-17 2014-04-01 Qualcomm Incorporated Grade of service (GoS) differentiation in a wireless communication network
US8139528B2 (en) * 2008-01-28 2012-03-20 Qualcomm Incorporated Adaptive transmission of resource utilization messages
US8867530B2 (en) * 2008-04-22 2014-10-21 Telefonaktiebolaget L M Ericsson (Publ) Signaling gateway for routing signaling messages
US8223641B2 (en) * 2008-07-28 2012-07-17 Cellco Partnership Dynamic setting of optimal buffer sizes in IP networks
US8031607B2 (en) * 2009-01-29 2011-10-04 Alcatel Lucent Implementation of internet protocol header compression with traffic management quality of service
CN101895403B (zh) * 2009-05-22 2013-08-28 华为技术有限公司 一种数据传送方法、装置和通信系统
US8780740B2 (en) * 2010-05-06 2014-07-15 Qualcomm Incorporated System and method for controlling downlink packet latency
KR20120010089A (ko) * 2010-07-20 2012-02-02 삼성전자주식회사 Ｈｔｔｐ 기반의 멀티미디어 스트리밍 서비스의 품질 향상을 위한 방법 및 장치
US20120114112A1 (en) * 2010-10-04 2012-05-10 Jon Rauschenberger Call center with federated communications
WO2012064069A2 (ko) * 2010-11-08 2012-05-18 엘지전자 주식회사 전송 파워 보고 방법 및 장치
JP5706297B2 (ja) * 2011-02-18 2015-04-22 アラクサラネットワークス株式会社 パケット転送装置及びＱｏＳ制御回路の電力供給制御方法
US20120233315A1 (en) * 2011-03-11 2012-09-13 Hoffman Jason A Systems and methods for sizing resources in a cloud-based environment
CN102739507B (zh) * 2011-04-14 2015-04-15 中国科学院声学研究所 一种承载状态感知的路由器及其业务流带宽分配方法
US9521205B1 (en) * 2011-08-01 2016-12-13 Google Inc. Analyzing changes in web analytics metrics
CN103139104B (zh) * 2011-12-05 2017-02-08 深圳迈瑞生物医疗电子股份有限公司 网络传输服务级别调整方法、数据终端和网络服务器
US9584385B2 (en) * 2012-01-19 2017-02-28 Comcast Cable Communications, Llc Adaptive buffer control
EP2728827A1 (en) * 2012-10-31 2014-05-07 British Telecommunications public limited company Communications network using a tunnel to connect two network nodes
US9191329B2 (en) * 2013-01-24 2015-11-17 Comcast Cable Communications, Llc Network latency optimization
US9485326B1 (en) * 2013-04-01 2016-11-01 Marvell Israel (M.I.S.L) Ltd. Scalable multi-client scheduling
US9282052B2 (en) * 2013-05-29 2016-03-08 Verizon Patent And Licensing Inc. End-to-end quality of service optimization and reporting
US9450881B2 (en) * 2013-07-09 2016-09-20 Intel Corporation Method and system for traffic metering to limit a received packet rate
CN104584627B (zh) * 2013-08-23 2018-05-18 华为技术有限公司 一种网络资源控制方法、装置和系统
US9882818B2 (en) * 2013-09-30 2018-01-30 Apple Inc. Adjusting a jitter buffer based on inter arrival jitter
US9191282B2 (en) * 2013-11-29 2015-11-17 Verizon Patent And Licensing Inc. Service level agreement (SLA) based provisioning and management
US9686204B2 (en) * 2014-02-05 2017-06-20 Verizon Patent And Licensing Inc. Capacity management based on backlog information
CA2882446A1 (en) * 2014-02-21 2015-08-21 Coho Data, Inc. Methods, systems and devices for parallel network interface data structures with differential data storage service capabilities
WO2015192319A1 (zh) * 2014-06-17 2015-12-23 华为技术有限公司 软件定义网络中识别攻击流的方法、装置以及设备
US9841325B2 (en) * 2014-08-29 2017-12-12 Oracle International Corporation High accuracy, compact on-chip temperature sensor
US10123323B2 (en) * 2014-10-24 2018-11-06 Qualcomm Incorporated Dynamic uplink/downlink frame structure for enhanced component carriers
EP3076616B1 (en) * 2015-03-31 2017-04-12 Mitsubishi Electric R&D Centre Europe B.V. Method for data traffic management and network node for implementing the same
US10798623B2 (en) * 2015-06-26 2020-10-06 Apple Inc. Method for processing radio signals and mobile terminal device
US10721147B2 (en) * 2015-11-05 2020-07-21 Cecelumen, Llc Methods and apparatus for associating content with one or more wireless signals and/or controlling access to content associated with such signals
US10110446B2 (en) * 2015-12-07 2018-10-23 Bank Of America Corporation Messaging queue spinning engine
KR20170100109A (ko) * 2016-02-24 2017-09-04 한국전자통신연구원 이동 통신 시스템에서의 버퍼 상태 보고 방법 및 그 장치
US9935888B2 (en) * 2016-05-02 2018-04-03 Visa International Service Association System and method for latency-based queuing
CN110995773B (zh) * 2016-05-24 2021-01-05 华为技术有限公司 QoS控制方法及设备
US11240697B2 (en) * 2016-08-11 2022-02-01 Lg Electronics Inc. Aperiodic CSI reporting method based on aperiodic CSI-RS in wireless communication system, and device therefor
US10972525B2 (en) * 2016-08-15 2021-04-06 Comcast Cable Communications, Llc Targeted buffer priority management
US10498612B2 (en) * 2016-09-27 2019-12-03 Mellanox Technologies Tlv Ltd. Multi-stage selective mirroring
US10148561B2 (en) * 2016-12-06 2018-12-04 At&T Intellectual Property I, L.P. Enhanced quality of service in software-defined networking-based connectionless mobility architecture
US10194446B2 (en) * 2017-02-07 2019-01-29 Qualcomm Incorporated Techniques and apparatuses for concurrent resource usage for WWAN and WLAN
CN108882367B (zh) * 2017-05-11 2020-10-30 维沃移动通信有限公司 数据传输的控制方法及装置
WO2018211508A1 (en) * 2017-05-17 2018-11-22 Cellwize Wireless Technologies Ltd. Method and system of managing radio connectivity of a vehicle
WO2019037846A1 (en) * 2017-08-23 2019-02-28 NEC Laboratories Europe GmbH SERVICE LEVEL MONITORING MONITORING METHOD IN A SOFTWARE DEFINED NETWORK AND CORRESPONDING SOFTWARE DEFINED NETWORK
US20190089760A1 (en) * 2017-09-20 2019-03-21 Junshan Zhang Systems and methods for real-time content creation and sharing in a decentralized network
WO2019069905A1 (ja) * 2017-10-04 2019-04-11 Ｎｅｃソリューションイノベータ株式会社 情報処理装置、制御方法、及びプログラム
CN109787801B (zh) * 2017-11-15 2022-01-21 华为技术有限公司 一种网络服务管理方法、装置和系统
KR102505060B1 (ko) * 2018-01-12 2023-03-02 삼성전자주식회사 무선 통신 시스템에서 스케줄링을 위한 장치 및 방법
CN112073216A (zh) * 2018-02-05 2020-12-11 华为技术有限公司 获取链路质量的方法和装置
US10798589B2 (en) * 2018-02-22 2020-10-06 FG Innovation Company Limited Methods and devices for measurement reporting in beam operations
CN110267310B (zh) * 2018-03-12 2022-04-22 中兴通讯股份有限公司 链路容量调整方法及装置、系统、控制器、网络节点
EP3565218B1 (en) * 2018-04-30 2023-09-27 Hewlett Packard Enterprise Development LP System and method of decentralized management of multi-owner nodes using blockchain
EP3791543B1 (en) * 2018-06-06 2023-03-01 Huawei Technologies Co., Ltd. Packet programmable flow telemetry profiling and analytics
US10764113B2 (en) * 2018-07-05 2020-09-01 At&T Intellectual Property I, L.P. Self-adjusting control loop
CN110958704B (zh) * 2018-09-26 2022-02-22 维沃移动通信有限公司 一种资源调度方法和装置
US12022495B2 (en) * 2018-09-28 2024-06-25 Sony Group Corporation Method and apparatus for adaptive bandwidth operation
US11196678B2 (en) * 2018-10-25 2021-12-07 Tesla, Inc. QOS manager for system on a chip communications
US11349545B2 (en) * 2018-11-02 2022-05-31 Apple Inc. Beam management without beam correspondence
CN109547362A (zh) * 2019-01-16 2019-03-29 郑州云海信息技术有限公司 一种队列调度方法、系统及电子设备和存储介质
JP7231749B2 (ja) * 2019-02-03 2023-03-01 華為技術有限公司 パケットスケジューリング方法、スケジューラ、ネットワーク装置及びネットワークシステム
US11664959B2 (en) * 2019-03-29 2023-05-30 Nokia Technologies Oy Dynamic configuration of user equipment duplexing mode
US20200334615A1 (en) * 2019-04-22 2020-10-22 Google Llc Predicting Business-Agnostic Contact Center Expected Wait Times With Deep Neural Networks
US11632338B2 (en) * 2019-07-03 2023-04-18 Vmware, Inc. Multi-tenant resource management in a gateway
US11240294B2 (en) * 2019-08-23 2022-02-01 Samsung Electronics Co., Ltd. Systems and methods for spike detection and load balancing resource management
US11729243B2 (en) * 2019-09-20 2023-08-15 Intel Corporation Dash-based streaming of point cloud content based on recommended viewports
US10817424B1 (en) * 2019-12-20 2020-10-27 Cloudflare, Inc. Using post-cache edge computing to re-populate dynamic content in cached content
CN113162790B (zh) * 2020-01-22 2023-10-03 华为技术有限公司 调整服务等级的方法、装置、设备及存储介质

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1918937A (zh) * 2004-02-03 2007-02-21 诺基亚公司 提供端对端服务质量的方法和设备
US20080247357A1 (en) * 2005-11-18 2008-10-09 Sk Telecom Co., Ltd. Method For Adaptive Delay Threshold-Based Priority Queueing Scheme For Packet Scheduling in Mobile Broadband Wireless Access System
CN109245936A (zh) * 2012-09-06 2019-01-18 华为技术有限公司 控制网络传输时延的方法、服务质量控制实体和通信设备

Also Published As

Publication number Publication date
CN113162789A (zh) 2021-07-23
JP2023511918A (ja) 2023-03-23
US20220377016A1 (en) 2022-11-24
JP7467645B2 (ja) 2024-04-15
US20220407808A1 (en) 2022-12-22
JP2023511916A (ja) 2023-03-23
EP4080829A1 (en) 2022-10-26
EP4084409A4 (en) 2023-06-28
KR20220126295A (ko) 2022-09-15
KR102704501B1 (ko) 2024-09-06
JP7531594B2 (ja) 2024-08-09
EP4080829A4 (en) 2023-06-28
EP4084409A1 (en) 2022-11-02
KR20220123686A (ko) 2022-09-08
CN113162790B (zh) 2023-10-03
KR102693696B1 (ko) 2024-08-08
WO2021148020A1 (zh) 2021-07-29
CN113162790A (zh) 2021-07-23

Similar Documents

Publication Publication Date Title
WO2021147368A1 (zh) 调整服务等级的方法、装置、设备、系统及存储介质
US11316795B2 (en) Network flow control method and network device
CN112671656B (zh) 一种对网络进行配置的方法和设备
US7948883B1 (en) Applying router quality of service on a cable modem interface on a per-service-flow basis
JP4619584B2 (ja) パケット交換網のルータにおいてパケットをスケジュール設定するための方法
US7027457B1 (en) Method and apparatus for providing differentiated Quality-of-Service guarantees in scalable packet switches
US9887923B2 (en) Opportunistic wireless resource utilization using dynamic traffic shaping
EP1345366A2 (en) Shared weighted fair queuing (WFQ) shaper
JP2006513612A (ja) ネットワーク通信における資源配分を実施するためのシステム及び方法
Irazabal et al. Dynamic buffer sizing and pacing as enablers of 5G low-latency services
EP3395023B1 (en) Dynamically optimized queue in data routing
WO2002098047A2 (en) System and method for providing optimum bandwidth utilization
US11463370B2 (en) Scalable deterministic services in packet networks
WO2022135202A1 (zh) 业务流的调度方法、装置及系统
JP7193787B2 (ja) 通信システム、ブリッジ装置、通信方法、及びプログラム
US12028265B2 (en) Software-defined guaranteed-latency networking
WO2022246710A1 (zh) 一种控制数据流传输的方法及通信装置
CN115550161A (zh) 参数配置的方法、装置、控制器、通信设备和通信系统
CN118740752A (zh) 流量控制方法和装置
CN113810314A (zh) 数据调度方法、设备和存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20915290

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022544401

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20227028085

Country of ref document: KR

Kind code of ref document: A

Ref document number: 2020915290

Country of ref document: EP

Effective date: 20220728

NENP Non-entry into the national phase

Ref country code: DE