WO2021120783A1 - 一种报文处理方法及装置 - Google Patents

一种报文处理方法及装置 Download PDF

Info

Publication number
WO2021120783A1
WO2021120783A1 PCT/CN2020/119154 CN2020119154W WO2021120783A1 WO 2021120783 A1 WO2021120783 A1 WO 2021120783A1 CN 2020119154 W CN2020119154 W CN 2020119154W WO 2021120783 A1 WO2021120783 A1 WO 2021120783A1
Authority
WO
WIPO (PCT)
Prior art keywords
message
packet
priority
flow
parameter
Prior art date
Application number
PCT/CN2020/119154
Other languages
English (en)
French (fr)
Inventor
白宇
王小忠
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
Publication of WO2021120783A1 publication Critical patent/WO2021120783A1/zh

Links

Images

Classifications

    • 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
    • H04L47/2433—Allocation of priorities to traffic types
    • 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
    • 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
    • H04L47/2441—Traffic characterised by specific attributes, e.g. priority or QoS relying on flow classification, e.g. using integrated services [IntServ]
    • 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
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/62—Queue scheduling characterised by scheduling criteria
    • H04L47/625—Queue scheduling characterised by scheduling criteria for service slots or service orders
    • H04L47/6275—Queue scheduling characterised by scheduling criteria for service slots or service orders based on priority

Definitions

  • This application relates to the field of data processing, and in particular to a message processing method and device.
  • Network equipment can forward packets, because the number of packets forwarded by the network equipment is relatively large, and the number of packets forwarded by the network equipment at the same time is limited. Therefore, for the message to be forwarded, the network device can determine the priority of the message, and forward the message according to the priority of the message. For example, the network device can forward packets according to the order of priority from high to low. It can be understood that, for a message with a higher priority, the corresponding transmission delay is relatively low, and for a message with a lower priority, the corresponding transmission delay may be relatively high.
  • the method of determining the message priority in the traditional technology cannot meet the transmission delay requirements of some messages.
  • the embodiment of the present application provides a message processing method to improve the transmission delay of the message.
  • the embodiments of the present application provide a message processing method, which can be applied to network devices such as routers and switches. Specifically, for a packet flow that is not a small flow, its requirements for transmission delay can be reflected in the characteristic parameters of the packet flow. Therefore, in the embodiment of the present application, the network device obtains the first packet and can determine the packet flow to which the first packet belongs. When the number of consecutively obtained packets belonging to the packet flow is greater than or equal to the first threshold, That is, when the packet flow is not a small flow, the characteristic parameter of the packet flow can be further determined according to the acquired first packet. Among the continuously acquired packets belonging to the packet flow, the time interval between adjacent packets is less than or equal to the first time threshold.
  • the priority of the first message can be determined according to the characteristic parameters of the message flow. For example, the characteristic parameter of the packet flow indicates that the packet flow requires a relatively low transmission delay, and the priority of the first packet can be determined to be a higher priority; for another example, the characteristic parameter of the packet flow indicates that the packet flow is allowed to be transmitted If the delay is relatively high, it can be determined that the priority of the first message is low priority, and so on. It can be seen that, using the solution of the embodiment of the present application, for the first packet whose packet flow is not a small flow, it is not directly determined as the low priority of the first packet as in the traditional technology. Instead, the priority of the first packet is determined in combination with the characteristic parameters of the packet flow to which the first packet belongs, so that the determined priority can meet the transmission delay requirement of the first packet.
  • these packet streams can be divided into multiple categories according to the requirements of the packet stream for transmission delay.
  • message streams that have basically the same level of transmission delay requirements can be divided into one category.
  • the priority of the message can be determined according to the category of the message flow, where one category can correspond to one priority.
  • packet streams of the same category have basically the same level of requirements for transmission delay, and the characteristic parameters of the packet stream can reflect the requirements of the packet stream for transmission delay. Therefore, for one or more packet streams belonging to a certain category, the characteristic parameters of the one or more packet streams can be considered to be within a certain parameter range.
  • the network device may, for example, predetermine the mapping relationship between the characteristic parameter range and the message priority, where the characteristic parameter range may reflect the type of the message flow.
  • the network device may further determine the first parameter range to which the characteristic parameters of the packet flow belong. Then, the network device can then determine the mapping relationship between the predetermined characteristic parameter range and the priority of the packet. , The message priority corresponding to the first parameter range is determined as the priority of the first message.
  • the network device may clear the aforementioned number of packets belonging to the packet flow that are continuously acquired.
  • the network device can also clear the characteristic parameters of the packet flow according to the aging period corresponding to the characteristic parameters. For the message flow to which the first message belongs, if the number of continuously acquired messages belonging to the message flow has been cleared to zero before the first message is acquired, then after the first message is acquired, the message The number of messages will be counted again from 1, which will affect the result of determining whether the message stream is a small stream, and accordingly, will affect the result of determining the priority of the first message.
  • the characteristic parameters of the packet flow will be recalculated after the first packet is obtained, which will also affect the determination of the first packet.
  • the result of the priority of the text It is understandable that if the priority of the first packet is determined incorrectly, it may cause the packet flow to which the first packet belongs to be out of order.
  • the so-called out-of-sequence message flow refers to the order in which the network equipment forwards the messages, which is inconsistent with the order in which the network equipment obtains the messages.
  • the first parameter may be stored in the network device, and the value of the first parameter is used to indicate the priority of the second packet.
  • the network device may determine the priority of the first packet in combination with the value of the first parameter and the characteristic parameter of the packet flow.
  • the network device may also modify the value of the aforementioned first parameter, so that the modified value of the first parameter indicates the priority of the first packet.
  • the network device acquires another message belonging to the message flow again, it can determine the priority of the aforementioned other messages by combining the value of the modified first parameter to reduce the number of messages. Flow out of order.
  • the first message may include a first field indicating the priority of the first message.
  • the value of the first field included in the first packet acquired by the network device may be inconsistent with the priority of the first packet determined according to the characteristic parameter of the packet flow. Therefore, in this embodiment of the present application, if the priority of the first packet determined by the network device does not match the priority of the packet indicated by the value of the first field in the first packet, the network device can perform a check on the first field. Modify so that the value of the modified first field indicates the priority of the first packet determined according to the characteristic parameter of the packet flow.
  • the network device may forward the first packet according to the priority of the first packet.
  • the characteristic parameter of the packet flow may also be sent to the display device ,
  • the display device displays the received characteristic parameters. In this way, the user can analyze the message flow through the content displayed by the display device.
  • the aforementioned characteristic parameter range may be issued by the controller to the network device.
  • the controller can also adjust the range of characteristic parameters of the message flow according to the characteristic parameters of the message flow. Specifically, after determining the characteristic parameter of the packet flow, the network device may send the characteristic parameter to the controller. It is understandable that the controller can receive the characteristic parameters of multiple packet streams, because each time the network device determines the characteristic parameter of a packet stream, the characteristic parameter can be sent to the controller. The controller can classify according to the acquired characteristic parameters of multiple message streams. The network device can then adjust the range of the aforementioned characteristic parameters according to the classification result. After the network device obtains the adjusted characteristic parameter, if the network device continues to obtain the third packet belonging to the packet flow, the priority of the third packet may be determined according to the adjusted characteristic parameter range.
  • the aforementioned range of characteristic parameters and the preset time period may affect the result of determining the priority of the first packet, resulting in that the forwarding delay of the first packet cannot satisfy the first packet.
  • the transmission delay requirement of a message if the result of the network device determining the message priority is relatively accurate, the actual delay of each queue in the network device is less than or equal to the upper limit of the delay corresponding to each queue. If there is a queue with an actual delay greater than its corresponding upper delay limit in the network device, it means that the aforementioned characteristic parameter range or the value of the preset time period is unreasonable.
  • the network device may also send the actual delay of the target queue to the controller, and the controller adjusts the target parameter according to the delay of the target queue and the upper limit of the target queue, so that The packet priority determined by the adjusted target parameter can make the actual delay of the target queue smaller than the upper limit of the delay of the target queue.
  • the controller may adjust the target parameter when the actual delay of the target queue is greater than the upper limit of the delay.
  • the target parameter mentioned here may include any one or more of the characteristic parameter range of the packet flow and the preset time period. After the controller adjusts the target parameter, it can send the adjusted target parameter to the network device. In this way, if the network device continues to obtain the third packet belonging to the packet flow, it can be based on the adjusted target The parameter determines the priority of the third message.
  • the characteristic parameter of the packet flow may be one or more parameters that can reflect the transmission delay requirement of the packet flow.
  • Some services that allow relatively high transmission delays such as file download services, will generate relatively large data traffic.
  • the servers that process the services In order to reduce the central processing unit (CPU) overhead required to process these data traffic, the servers that process the services often A large number of messages will be sent in batches in stages, which leads to a large burst of the message flow corresponding to the service, and the maximum message time interval is also large.
  • the message time interval mentioned here refers to the time interval of adjacent messages.
  • the generated traffic is relatively small compared to services such as video downloads, so the corresponding bursts are relatively small.
  • the servers that process the services often send packets in real time and continuously, which results in a relatively small maximum packet time interval for the packet flow corresponding to the service.
  • the number of packets included in the burst of a packet flow and the maximum packet time interval can both reflect the requirements of the corresponding service of the packet flow on the transmission delay. Therefore, the characteristic parameters of the packet flow can include the burst of the packet flow. The number of packets.
  • the characteristic parameter of the message flow may also include the maximum message interval.
  • the rate of a packet stream with a relatively low transmission delay is generally required to be relatively stable, while the rate of a packet stream with a relatively high transmission delay is allowed to fluctuate relatively greatly.
  • the characteristic parameters of the message flow may also include the rate fluctuation coefficient of the message flow.
  • the maximum instantaneous rate of a packet stream with a relatively low transmission delay is generally not very large, while the maximum instantaneous rate of a packet stream with a relatively high transmission delay is allowed to be very large.
  • the transmission delay requirement of the message flow can also be determined in combination with the instantaneous rate of the message flow. Therefore, the characteristic parameter of the message flow may also include the maximum instantaneous rate of the message flow.
  • an embodiment of the present application provides a message processing device, the device including: an acquiring unit, a first determining unit, a second determining unit, and a third determining unit.
  • the obtaining unit is used for obtaining the first packet; the first determining unit is used for determining the packet flow to which the first packet belongs; the second determining unit is used for when the number of consecutively obtained packets belonging to the packet flow is greater than Or when it is equal to the first threshold, the characteristic parameter of the packet flow is determined according to the first packet, and among the continuously obtained packets belonging to the packet flow, the time interval between adjacent packets is less than or equal to The first time threshold; the third determining unit is configured to determine the priority of the first packet according to the characteristic parameter of the packet flow.
  • the third determining unit is specifically configured to: when the characteristic parameter of the packet flow is within the first parameter range, according to the predetermined mapping relationship between the characteristic parameter range and the packet priority,
  • the message priority corresponding to the first parameter range is determined as the priority of the first message, and the mapping relationship between the characteristic parameter range and the message priority includes: the first parameter range and the first parameter range.
  • the third determining unit is specifically configured to: obtain the value of the first parameter, and the value of the first parameter is used to indicate the priority of the second message, and the second message is related to the priority of the second message.
  • the first message belongs to the same message flow, and the second message is the previous message of the first message; the value of the first parameter and the characteristic parameter of the message flow are used to determine the The priority of the first message.
  • the device further includes a first modification unit configured to modify the value of the first parameter, and the modified value of the first parameter is used to indicate the first report. The priority of the text.
  • the device further includes a second modification unit configured to modify the value of the first field in the first message, and the modified value of the first field indicates the The priority of the first message.
  • the second modification unit the device further includes a forwarding unit, and the forwarding unit is configured to forward the first message according to the priority of the first message.
  • the device further includes a first sending unit, and the first sending unit is configured to send the characteristic parameters of the packet stream to a display device for display.
  • the device further includes a second sending unit and an adjustment unit.
  • the second sending unit is used to send the characteristic parameters of the packet flow to the controller;
  • the adjustment unit is used to adjust the corresponding parameters according to the classification results of the characteristic parameters of the multiple packet flows obtained from the controller.
  • the characteristic parameter range of the message flow, the adjusted characteristic parameter range corresponding to the message flow is used to determine the priority of the third message, and the third message is the acquiring unit acquiring the first message For the messages obtained afterwards, the third message belongs to the message flow.
  • the device further includes a third sending unit.
  • the third sending unit is used to send the actual delay of the target queue to the controller, so that the controller adjusts the target parameter when the actual delay of the target queue is greater than the upper limit of the target queue delay.
  • the latter target parameter is used to determine the priority of a third packet, which is a packet obtained after the network device obtains the first packet, and the third packet belongs to the A message flow; wherein: the target queue is a queue in the network device for storing messages to be forwarded; the target parameter includes any one or more of the following: the characteristic parameter range corresponding to the message flow and Each feature parameter corresponds to a preset time threshold, the first feature parameter is any one of the feature parameters, and the preset time threshold corresponding to the first feature parameter is used to indicate that the network device is in phase. When the time interval of adjacent messages is greater than the preset time threshold, the first characteristic parameter is cleared.
  • the characteristic parameters of the packet flow include any one or more of the following: the number of packets included in the burst of the packet flow, the maximum packet time interval of the packet flow, The message time interval is the time interval of adjacent messages, the rate fluctuation coefficient of the message flow, and the maximum instantaneous rate of the message flow.
  • an embodiment of the present application provides a device.
  • the device includes a processor and a memory.
  • the memory is used to store instructions or computer programs.
  • the processor is configured to execute the instruction or computer program in the memory, and execute the method described in any one of the first aspect above.
  • an embodiment of the present application provides a computer-readable storage medium, including instructions or computer programs, which when run on a computer, cause the computer to execute the method described in any one of the first aspects above.
  • the embodiments of the present application provide a computer program product containing instructions or computer programs, which when run on a computer, cause the computer to execute the method described in any one of the first aspects above.
  • Figure 1 is a schematic diagram of an exemplary application scenario
  • FIG. 2 is a schematic flowchart of a message processing method provided by an embodiment of this application.
  • FIG. 3 is a schematic diagram of a method for determining a burst of a packet flow according to an embodiment of the application
  • FIG. 4 is a schematic structural diagram of a message processing apparatus provided by an embodiment of the application.
  • FIG. 5 is a schematic structural diagram of a message processing device provided by an embodiment of this application.
  • FIG. 6a is a schematic diagram of a network architecture provided by an embodiment of this application.
  • FIG. 6b is a schematic diagram of a network architecture of a data center network provided by an embodiment of this application.
  • Fig. 7a is a schematic diagram of an exemplary application scenario provided by an embodiment of the application.
  • Fig. 7b is a schematic diagram of an exemplary application scenario provided by an embodiment of the application.
  • the embodiment of the present application provides a message processing method, which is used to improve the transmission delay of the message.
  • Figure 1 is a schematic diagram of an exemplary application scenario. Before introducing the scenario shown in FIG. 1, the concept of "message flow" involved in the following embodiments is first introduced.
  • the information can be a two-tuple, a five-tuple, or a seven-tuple.
  • the two-tuple includes a source Internet Protocol (IP) address and a destination IP address.
  • IP Internet Protocol
  • the five-tuple includes source IP address, source port number, destination IP address, destination port number, and transport layer protocol number.
  • the seven-tuple includes a source media access control (MAC) address, a destination MAC address, a source IP address, a destination IP address, a protocol number, a source port number, and a destination port number.
  • MAC media access control
  • the continuous message can also be referred to as a message stream. Therefore, the message flow to which the message belongs can be determined by the information carried in the message, such as a two-tuple, a five-tuple, a seven-tuple, or other information that can be used to distinguish the message flow.
  • the device 101 may obtain a message from the device 102 and forward the message to the device 103.
  • the embodiment of the present application does not specifically limit the specific forms of the device 101, the device 102, and the device 103.
  • the device 101 may be a network device such as a router or a switch.
  • the device 102 may be a network device such as a router and a switch, or may be a device such as a server or a terminal device.
  • the device 103 may be a network device such as a router or a switch, or may be a device such as a server or a terminal device.
  • the number of packets obtained by the device 101 is relatively large, but the packet forwarding capability of the device 101 is limited, that is, the number of packets forwarded by the device 101 at the same time is limited.
  • the device 101 can determine each packet. The priority of the message, and the message is forwarded according to the priority of each message.
  • FIG. 1 is only shown to facilitate understanding of the application scenarios of the embodiments of the present application, and it does not constitute a limitation to the embodiments of the present application.
  • the message obtained by the device 101 may correspond to multiple services, such as a video-on-demand service or a voice service.
  • Different services have different requirements for flow completion time (FCT).
  • FCT flow completion time
  • a voice call service requires a relatively short completion time for a message flow
  • a video-on-demand service allows a relatively long flow completion time.
  • the flow completion time of the message flow may be reflected by the transmission delay of the messages included in the message flow.
  • Packets of different services have different requirements for transmission delay.
  • the voice call service requires relatively low transmission delay
  • the video-on-demand service allows relatively high transmission delay.
  • the packet flow corresponding to the service requiring relatively low transmission delay is generally a small flow.
  • the so-called small flow refers to a packet flow with a relatively low rate or a packet flow with a relatively small number of packets. Therefore, in the traditional technology, the method of determining the priority of a message is relatively simple, that is: if the message flow to which the message belongs is a small flow, the priority of the message is high priority, and if the message to which the message belongs is a small flow, the priority of the message is high priority. If the flow is not a small flow, the priority of the packet is low priority.
  • the corresponding message stream is not a small stream, the service also requires a relatively low transmission delay.
  • services such as video conferencing services and virtual reality (virtual reality, VR)
  • VR virtual reality
  • the corresponding message streams are not small streams, such services also require relatively low transmission delay.
  • the traditional technology for determining the priority of the message will determine the priority of the message of this type of service as a low priority, so that the determined priority cannot meet the transmission delay requirements of the message. Furthermore, it will affect the quality of service (QOS) of the business.
  • QOS quality of service
  • the embodiment of the present application provides a message processing method, which is used to determine the priority of the message, and the determined priority of the message can meet the transmission delay requirement of the message. Specifically, using the solution of the embodiment of the present application, when the packet stream to which the first packet to be forwarded belongs is not a small stream, the determined priority of the first packet can satisfy the first packet pair transmission time. Extension of the request.
  • the message processing method provided by the embodiments of the present application will be introduced below with reference to the accompanying drawings.
  • FIG. 2 is a schematic flowchart of a message processing method provided by an embodiment of the application.
  • the message processing method shown in FIG. 2 can be implemented through the following S101-S103, for example.
  • a network device is taken as an execution subject as an example.
  • the network device obtains the first packet, and determines the packet flow to which the first packet belongs.
  • the network device may be an intermediate node that forwards the first packet.
  • the network device may receive the first message from other devices, and after obtaining the first message, the network device may forward the first message to other devices.
  • the network device mentioned in the embodiment of this application may be, for example, the device 101 shown in FIG. 1.
  • the network device after the network device obtains the first packet, it can further determine the packet flow to which the first packet belongs, and determine the packet flow of the first packet according to the packet flow to which the first packet belongs. priority. Specifically, the network device may determine the packet flow to which the first packet belongs according to the 2-tuple, 5-tuple, or 7-tuple of the first packet.
  • S102 When the number of continuously obtained packets belonging to the packet flow is greater than or equal to the first threshold, the network device determines the characteristic parameters of the packet flow according to the first packet, and the continuously obtained packets belonging to the packet flow In the messages, the time interval between adjacent messages is less than or equal to the first time threshold.
  • S103 The network device determines the priority of the first packet according to the characteristic parameter of the packet flow.
  • the network device may further determine whether the number of continuously obtained packets belonging to the packet flow is less than or equal to the first threshold to determine the packet to which the first packet belongs Whether the flow is a small flow, wherein, among the continuously obtained packets belonging to the packet flow, the time interval between adjacent packets is less than or equal to the first time threshold.
  • the specific value of the first time threshold may be determined according to actual conditions, and the first time threshold may be, for example, 500 milliseconds.
  • the priority of the first message is high priority.
  • the priority of the first packet is further determined according to the requirement of the packet flow for transmission delay. Specifically, the characteristic parameters of the packet flow may reflect the requirements of the packet flow on the transmission delay. Therefore, the characteristic parameters of the packet flow may be further determined, thereby further determining the priority of the first packet.
  • the characteristic parameter of the packet flow may be one or more parameters that can reflect the transmission delay requirement of the packet flow.
  • Some services that allow relatively high transmission delays, such as file download services, will generate relatively large data traffic.
  • CPU central processing unit
  • the servers that process the services often A large number of messages are sent in batches in stages, which results in a larger burst of the message flow corresponding to the service, and a larger maximum message interval.
  • the message time interval mentioned here refers to the time interval of adjacent messages.
  • the generated traffic is relatively small compared to services such as video downloads, so the corresponding bursts are relatively small.
  • the servers that process the services often send packets in real time and continuously, which results in a relatively small maximum packet time interval for the packet flow corresponding to the service. .
  • FIG. 3 is a schematic diagram of a method for determining a burst of a packet flow according to an embodiment of the application. As shown in Figure 3, the horizontal axis represents the time to obtain the packet, and the vertical axis represents the number of packets included in the burst, that is, the burst size. The small square below the horizontal axis represents messages.
  • the larger burst mentioned above refers to the larger number of packets included in the burst, and the smaller burst refers to the smaller number of packets included in the burst.
  • the characteristic parameter of the packet flow may include the number of packets included in the burst of the packet flow.
  • the characteristic parameter of the packet flow may further include the maximum packet time interval.
  • g represents the maximum packet time interval of a packet flow
  • the network device obtains a new packet of the packet flow
  • the t1 mentioned here can be understood as the aging period corresponding to the maximum message interval.
  • the characteristic parameter of the message flow may also include the rate fluctuation coefficient of the message flow.
  • the rate fluctuation coefficient may be, for example, the variance of the rate of the message flow, or It is the standard deviation of the rate of the message flow, and can also be the ratio of the standard deviation of the message flow to the average rate of the message flow.
  • the maximum instantaneous rate of a packet stream with a relatively low transmission delay is generally not very large, while the maximum instantaneous rate of a packet stream with a relatively high transmission delay is allowed to be very large.
  • the transmission delay requirement of the message flow can also be determined in combination with the instantaneous rate of the message flow. Therefore, in another implementation manner of the embodiment of the present application, the characteristic parameter of the packet flow may also include the maximum instantaneous rate of the packet flow.
  • the rate of the message flow can be determined as follows: Determine the number of bytes of the message included in the message flow in the time period T1, and divide the number of bytes by T1 to obtain the rate of the message flow. It is understood that if the value of T1 is small, the instantaneous rate of the packet flow can be calculated, and if the T1 is relatively large, the average rate of the packet flow can be calculated. After the rate of the message flow is calculated, the maximum instantaneous rate of the message flow and the rate fluctuation coefficient of the message flow can be further obtained.
  • the network device may pre-store the mapping relationship between the characteristic parameter and the message priority.
  • the network device determines the characteristic parameter of the packet flow to which the first packet belongs, and determines the priority of the first packet according to the pre-stored mapping relationship.
  • the mapping relationship includes at least the mapping relationship between the characteristic parameter determined in S102 and the priority of the first packet determined in S103.
  • message streams that are not small streams these message streams can be divided into multiple categories according to the requirements of the message stream for transmission delay.
  • message streams that have basically the same level of transmission delay requirements can be divided into one category.
  • the priority of the message can be determined according to the category of the message flow, where one category can correspond to one priority.
  • Packet streams of the same category have basically the same level of requirements for transmission delay, and the characteristic parameters of the packet stream can reflect the requirements of the packet stream for transmission delay. Therefore, for one or more packet streams belonging to a certain category, the characteristic parameters of the one or more packet streams can be considered to be within a certain parameter range.
  • the network device may, for example, predetermine the mapping relationship between the characteristic parameter range and the message priority, where the characteristic parameter range may reflect the type of the message flow.
  • the network device may further determine the first parameter range to which the characteristic parameters of the packet flow belong. Then, the network device can then determine the mapping relationship between the predetermined characteristic parameter range and the priority of the packet. , The message priority corresponding to the first parameter range is determined as the priority of the first message.
  • the category of the message flow can be, for example, delay-sensitive, which means that the service corresponding to the message flow is more sensitive to delay, that is, the message flow requires a relatively low transmission delay;
  • the category of the message flow can also be delay-insensitive, which means that the service corresponding to the message flow is insensitive to delay, that is, the message flow allows a relatively high transmission delay.
  • the delay-sensitive category can also be subdivided into the delay-sensitive category, the delay-sensitive category, and the delay-low-sensitive category. List one by one.
  • the characteristic parameters of the message flow can include not only the number of messages included in the burst, but also the maximum message interval, and of course, the rate fluctuation coefficient and the maximum instantaneous rate mentioned above.
  • the packet flow category may include the delay-sensitive category, the delay-sensitive category, and the delay-insensitive category shown in Table 1, as well as other categories. For example, it may also include the delay-low sensitivity category. Wait, I won't list them one by one here.
  • the range of characteristic parameters is not limited to those shown in Table 2.
  • the message priority can be high, medium, low, etc. in addition to the high, medium, and low shown in Table 2.
  • Table 1 is only shown to illustrate the physical meaning represented by the characteristic parameter range.
  • the network device may not need to determine the type of the packet flow according to the mapping relationship shown in Table 1.
  • the network device can directly determine the priority of the first packet according to the mapping relationship shown in Table 2.
  • Characteristic parameter range category The number of packets included in the burst is less than or equal to 8 Delay High Sensitive Class The number of packets included in the burst is greater than 8 and less than or equal to 39 Delay-sensitive category The number of packets included in the burst is greater than 39 Delay insensitive
  • the maximum message interval is greater than or equal to 500 microseconds Delay insensitive
  • the maximum message interval is less than 500 microseconds Delay-sensitive category
  • Characteristic parameter range Message priority
  • the number of packets included in the burst is less than or equal to 8 Higher
  • the number of packets included in the burst is greater than 8 and less than or equal to 39 in
  • the number of packets included in the burst is greater than 39 low
  • the network device may have a storage device for storing The first flow table of the aforementioned number of messages.
  • the packet flow to which the first packet belongs may include multiple packets, and the first packet may be the first packet of the packet flow, or may not be the first packet of the packet flow.
  • the network device After the network device obtains the first packet, it can determine that the number of packets is 1, and store the number of packets in the first packet.
  • First class table When the first packet is the first packet of the packet flow, after the network device obtains the first packet, it can determine that the number of packets is 1, and store the number of packets in the first packet.
  • the number of packets corresponding to the network device before acquiring the first packet may have been stored in the aforementioned first flow table, assuming the packets stored in the first flow table If the number is N, after obtaining the first packet, the network device can determine whether the time interval between obtaining the first packet and obtaining the second packet is greater than the first time threshold. If it is greater than, the network device can store in the first flow table The number of packets is cleared. If it is less than or equal to, the network device can determine that the number of packets is N+1.
  • the first time threshold mentioned here may also be referred to as the aging period of the number of packets.
  • the second message mentioned here is the previous message of the first message, and the second message and the first message belong to the same message flow.
  • the network device when it stores the number of messages in the first flow table, for example, it may first perform a hash operation on the 2-tuple, quintuple, or seven-tuple of the first packet. , Obtain the corresponding message quantity storage address, and then obtain the value of the message quantity stored in the address, and add 1 to the value, or clear the value to zero.
  • a classic sketch technique can be used to determine the number. For example, it can be determined by using the count min sketch (CM sketch) technology in the sketch technology.
  • the network device may have a second flow table for storing characteristic parameters of the packet flow.
  • the second flow table may store the identifier of the packet flow and the characteristic parameters of the packet flow.
  • the identifier of the packet flow mentioned in the embodiment of the present application may be, for example, a two-tuple, a five-tuple, or a seven-tuple of the packet flow. Therefore, in an implementation manner of the embodiment of the present application, after the network device determines the packet flow to which the first packet belongs, it can further determine whether there is a combination of the aforementioned two-tuple, five-tuple, or seven-tuple in the second flow table. The characteristic parameter corresponding to the group. If it exists, the network device can also directly determine that the packet flow is not a small flow.
  • the network device after the network device obtains the first packet, it can first determine whether the characteristic parameter of the packet flow to which the first packet belongs exists in the second flow table. If so, the network device can directly determine the first packet.
  • the packet stream to which a packet belongs is not a small stream. In this case, before acquiring the first message, the message flow to which the first message belongs has been determined to be not a small flow.
  • the foregoing continuously acquired messages belonging to the message stream are the messages acquired before the first message is acquired.
  • the network device may further query the first flow table to determine the number of packets N included in the packet flow before acquiring the first packet, and then If the time interval between the network device acquiring the first packet and acquiring the second packet is less than or equal to the first time threshold, the network device may add 1 to the number N, and further determine whether N+1 is greater than or equal to the first time threshold. A threshold. If N+1 is greater than or equal to the first threshold, the network device can determine that the packet flow to which the first packet belongs is not a small flow. If N+1 is less than the first threshold, the network device can determine that the packet flow to which the first packet belongs is a small flow.
  • the network device clears the number N in the first flow table to zero, and determines that the packet flow to which the first packet belongs is small flow. In this case, before the first packet is obtained, the packet flow to which the first packet belongs is not determined to be not a small flow, and the aforementioned continuously obtained packets belonging to the packet flow include the first packet.
  • each characteristic parameter may correspond to a preset time interval.
  • the preset time interval corresponding to the first characteristic parameter is used to indicate to clear the first characteristic parameter of the packet stream. Specifically, when the time interval between adjacent packets of the packet flow is greater than the preset time interval, the first characteristic parameter of the packet flow is cleared.
  • the preset time interval may be obtained by the network device through a human-computer interaction interface, or may be sent to the network device by the controller, which is not specifically limited in the embodiment of the present application.
  • the embodiment of the present application does not specifically limit the specific value of the preset time threshold.
  • the specific value of the preset time threshold may be determined according to actual conditions. For example, the specific value of the preset time threshold may be 200 microseconds.
  • the preset time interval corresponding to the first characteristic parameter mentioned here may also be referred to as the aging period of the first characteristic parameter.
  • the aforementioned first threshold and the range of characteristic parameters may be issued by the controller to the network device, or of course, may also be obtained by the network device through a human-computer interaction interface, which is not specifically limited in the embodiment of the present application.
  • the priority of the first packet is determined in combination with the characteristic parameters of the packet flow to which the first packet belongs, so that the determined priority can meet the transmission delay requirement of the first packet.
  • the first message may include a first field indicating the priority of the first message.
  • the value of the first field included in the first packet obtained by the network device may be inconsistent with the priority of the first packet determined in S103.
  • the method for determining the priority of the message by the device that sends the first message to the network device is the same as that in the conventional technology, so the determined priority of the first message is the low priority. Therefore, in the embodiment of the present application, after the network device determines the priority of the first packet, if the determined priority of the first packet is different from the priority of the packet indicated by the value of the first field in the first packet If they match, the network device can modify the first field so that the value of the modified first field indicates the priority of the first packet determined in S103.
  • the embodiment of the present application does not specifically limit the first field, and the first field may be, for example, a differentiated services code point (DSCP) field.
  • DSCP differentiated services code point
  • the network device may forward the first packet according to the determined priority. Specifically, the network device may add the first message to the queue corresponding to the priority of the first message, so as to achieve the purpose of forwarding the first message according to the priority of the first message.
  • the network device forwards the first packet, if the network device that receives the first packet The device can correctly parse the value of the first field, and the network device that receives the first packet can determine the priority of the first packet according to the value of the first field, and further based on the priority of the first packet The first message is forwarded, so that in the forwarding process of the first message, its priority can meet the transmission delay requirement of the message flow of the first message.
  • the network device when the time interval between adjacent packets is greater than the first time threshold, can clear the aforementioned number of packets belonging to the packet flow that are continuously acquired. In addition, the network device can also clear the characteristic parameters of the packet flow according to the aging period corresponding to the characteristic parameters. For the message flow to which the first message belongs, if the number of the aforementioned messages has been cleared to zero before the first message is obtained, the number of messages will be counted from 1 again after the first message is obtained. It will affect the result of determining whether the packet flow is a small flow, and accordingly, it will affect the result of determining the priority of the first packet. In addition, if the characteristic parameters of the packet stream have been cleared before the first packet is obtained, the characteristic parameters of the packet stream will be recalculated after the first packet is obtained, which will also affect the determination of the first packet. The result of the priority of the text.
  • the message flow to which the first message belongs may be out of sequence.
  • the so-called out-of-sequence message flow refers to the order in which the network equipment forwards the messages, which is inconsistent with the order in which the network equipment obtains the messages.
  • the previous message of the first message is the second message, and the second message belongs to the same message flow as the first message.
  • the characteristic parameter of the message stream to which the first message belongs has been cleared. Assuming that before the first packet is obtained, the category corresponding to the packet flow is the delay-insensitive category, and correspondingly, the priority of the second packet is determined to be the low priority.
  • the characteristic parameters of the packet flow to which the first packet belongs will be recalculated, if determined according to the recalculated characteristic parameters
  • the priority of the first message is a higher priority. This causes the network device to forward the first message earlier than the time between forwarding the second message, which results in out-of-sequence of the message flow.
  • the first parameter may be stored in the network device, and the value of the first parameter is used to indicate the priority of the second packet.
  • the network device may determine the priority of the first packet in combination with the value of the first parameter and the characteristic parameter of the packet flow.
  • the first priority may be determined according to the value of the first parameter
  • the second priority may be determined according to the characteristic parameter of the packet flow
  • the lower priority of the first priority and the second priority may be determined. Determined as the priority of the first message. In this way, it can be ensured that the priority of the first message is not higher than the priority of the second message, thereby avoiding the disorder of the message flow.
  • the value of the first parameter indicates that the priority of the second packet is low priority, that is, before the first packet is obtained, the type of the packet flow is delay-insensitive, which is determined according to the value of the first parameter
  • the first priority is low priority.
  • the second priority determined according to the characteristic parameters of the message flow is a higher priority, so it can be determined that the priority of the first message is a low priority.
  • the reason why the second priority is a higher priority may be caused by the feature parameters of the packet stream that have been cleared before the first packet is obtained.
  • the value of the first parameter indicates that the priority of the second packet is a higher priority, that is, before the first packet is obtained, the type of the packet flow is delay-sensitive, which is determined according to the value of the first parameter
  • the first priority is the higher priority.
  • the second priority determined according to the characteristic parameters of the message flow is the low priority, so the priority of the first message can be determined to be the low priority. Because according to the characteristic parameters before the first packet is obtained, the type of the packet flow cannot be determined as delay-insensitive. After the first packet is obtained, the packet can be determined based on the characteristic parameters calculated on the first packet.
  • the class of the flow is delay-insensitive.
  • the number of packets included in the burst before the first packet is obtained is 39, so the type of the packet flow is determined to be delay sensitive, and after the first packet is obtained, the number of packets included in the burst is 40. Therefore, it can be determined that the type of the packet flow is delay-insensitive.
  • the network device may also modify the value of the aforementioned first parameter, so that the modified value of the first parameter indicates the priority of the first packet. It is understandable that after the value of the first parameter is modified, if the network device obtains another message belonging to the message flow again, it can determine the priority of the aforementioned other messages in combination with the value of the modified first parameter. Level to avoid disorder of message flow.
  • the value of the first parameter may be stored in the second flow table mentioned above, specifically, the packet flow to which the first packet belongs may be stored in the second flow table.
  • the identifier of the message flow may be a two-tuple, a five-tuple or a seven-tuple of the message flow.
  • the network device may also clear the value of the first parameter according to a certain period, for example, the first period, and release the data for storing the first parameter accordingly.
  • the memory space of the value When the first parameter is cleared, the identifier of the packet flow can also be deleted accordingly.
  • the specific value of the second period is not specifically limited in the embodiment of the present application.
  • the value of the first period may be, for example, a value that ensures that the packet flow will not be out of order.
  • the value of the first period may be 2 milliseconds.
  • the characteristic parameter of the packet flow may also be determined.
  • the parameters are sent to the display device, and the display device displays the received characteristic parameters.
  • the display device may display the aforementioned characteristic parameters in graphical forms such as bar graphs, line graphs, etc., for example. It is understandable that the user can analyze the message flow through the content displayed by the display device.
  • the network device can obtain the number of messages included in each burst of the message flow, and send the number of messages included in each burst.
  • the content displayed by the display device can be, in addition to the number of messages included in the burst, it can also be the total bytes of the messages included in the burst. number. Among them, the total number of message bytes included in the burst can be obtained by multiplying the average number of bytes of each message by the number of messages included in the burst. In this way, the user can determine the burst distribution of messages through the content displayed by the display device.
  • the display device may be a device with a display function, for example, the display device may be a terminal device.
  • the aforementioned characteristic parameter range may be issued by the controller to the network device.
  • the controller The range of characteristic parameters of the packet flow can also be adjusted according to the characteristic parameters of the packet flow.
  • the network device may send the characteristic parameter to the controller.
  • the controller can receive the characteristic parameters of multiple packet streams, because every time the network device determines the characteristic parameters of a packet stream, the characteristic parameters can be sent to the controller.
  • the controller can classify according to the acquired characteristic parameters of multiple message streams.
  • the network device can then adjust the range of the aforementioned characteristic parameters according to the classification result.
  • the embodiment of the present application does not specifically limit the classification algorithm used by the controller to classify the characteristic parameters, and the classification algorithm may be, for example, a classic support vector machine (SVM) algorithm.
  • SVM support vector machine
  • the controller divides the number of messages included in the burst of multiple message flows into three types.
  • the first type corresponds to the burst including
  • the number of packets corresponding to the second type is less than or equal to 9
  • the number of packets corresponding to the second type of burst is greater than 9 and less than or equal to 40
  • the number of packets corresponding to the third type is greater than 40.
  • the adjusted characteristic parameter range can be understood by referring to the mapping relationship between the characteristic parameter range and the message flow category shown in Table 3.
  • Characteristic parameter range category The number of packets included in the burst is less than or equal to 9 Delay High Sensitive Class The number of packets included in the burst is greater than 9 and less than or equal to 40 Delay-sensitive category The number of packets included in the burst is greater than 40 Delay insensitive
  • the network device can adjust the range of the characteristic parameter according to the classification result, and there may be two implementation methods.
  • the controller may adjust the characteristic parameter range, and after the controller adjusts the characteristic parameter range, the adjusted characteristic parameter range may be sent to the network device.
  • the controller may send the foregoing classification result to the network device, and the network device adjusts the characteristic parameter according to the classification result. After the network device obtains the adjusted characteristic parameter, if the network device continues to obtain the third packet belonging to the packet flow, the priority of the third packet may be determined according to the adjusted characteristic parameter range.
  • the aforementioned range of characteristic parameters and the preset time period may affect the result of determining the priority of the first packet, resulting in that the forwarding delay of the first packet cannot satisfy the transmission delay of the first packet. Delay requirements.
  • the result of the network device determining the message priority is relatively accurate, the actual delay of each queue in the network device is less than or equal to the upper limit of the delay corresponding to each queue. If there is a queue with an actual delay greater than its corresponding upper delay limit in the network device, it means that the aforementioned characteristic parameter range or the value of the preset time period is unreasonable.
  • the network device may also send the actual delay of the target queue to the controller, and the controller adjusts the target parameter according to the delay of the target queue and the upper limit of the target queue, so that The message priority determined by the adjusted target parameter can make the actual delay of the target queue smaller than the upper limit of the delay of the target queue.
  • the controller may adjust the target parameter when the actual delay of the target queue is greater than the upper limit of the delay.
  • the target parameter mentioned here may include any one or more of the aforementioned characteristic parameter range and preset time period.
  • the target queue may include one or more queues in the network device.
  • the target queue may include any one or more of queue a, queue b, and queue c mentioned above.
  • the upper limit of the delay of the target queue may be predetermined.
  • the upper limit of the delay may be an empirical value, or it may be obtained by the controller through statistics of the historical delay data of the target queue in the network device, which is not done in this embodiment of the application. Specific restrictions.
  • the actual delay of the target queue may be calculated by the network device, and the network device may determine the instantaneous queue delay of the target queue as the actual delay of the target queue.
  • the network device can obtain the message depth of the target queue, that is, the total number of message bytes included in the target queue, and obtain the instantaneous rate of the target queue, which can be obtained by dividing the message depth of the target queue by the instantaneous rate of the target queue The instantaneous queue delay of the target queue.
  • the controller after the controller adjusts the target parameter, it can send the adjusted target parameter to the network device. In this way, if the network device continues to obtain the third packet belonging to the packet flow, then The priority of the third message can be determined according to the adjusted target parameter.
  • the method for the controller to adjust the target parameters is illustrated with examples.
  • the target queue is the aforementioned queue b, and the type of the packet flow corresponding to the queue b is delay-sensitive.
  • the actual delay of the target queue is 2.3 milliseconds, and the upper limit of the delay of the target queue is 2.0 milliseconds. Since the actual delay is greater than the upper limit of the delay, it indicates that there are too many packets queued in queue b.
  • the reason for the excessive number of packets queued in queue b may be that the type of packet flow corresponding to some of the packets in queue b is non-delay sensitive.
  • the controller can adjust the preset time period corresponding to the number of bursts, that is, adjust the aging period of the number of packet streams included in the burst, for example, increase the aging period, thereby
  • the number of packet streams included in the bursts of the packet flow can fall within the parameter range corresponding to the delay-insensitive class.
  • the number of bursts included in the packet stream can be greater than 40 shown in Table 3, so that Some messages in queue b can be divided into queue c again, thereby reducing the number of messages queued in queue b, and further optimizing the result of determining message priority.
  • the type of the packet flow corresponding to the queue c mentioned here is the delay-insensitive type.
  • the controller can also adjust the range of characteristic parameters corresponding to the non-delay-sensitive class, for example, adjust the range of characteristic parameters corresponding to the non-delay-sensitive class from greater than 40 to greater than 38, so that some packets in queue b can be It is re-divided into queue c to slow down the number of messages queued in queue b, and further optimize the result of determining message priority.
  • the embodiment of the present application also provides a corresponding device, which is described below with reference to the accompanying drawings.
  • FIG. 4 is a schematic structural diagram of a message processing apparatus provided by an embodiment of the application.
  • the time message processing apparatus 400 shown in FIG. 4 can be used to execute the message processing method shown in FIG. 2.
  • the message processing apparatus 400 may include an acquiring unit 401, a first determining unit 402, a second determining unit 403, and a third determining unit 404.
  • the acquiring unit 401 is configured to acquire a first message; the first determining unit 402 is configured to determine the message stream to which the first message belongs; and the second determining unit 403 is configured to determine when the continuously acquired messages belong to the message stream.
  • the characteristic parameters of the packet flow are determined according to the first packets, and among the continuously acquired packets belonging to the packet flow, adjacent packets
  • the third determining unit 404 is specifically configured to: when the characteristic parameter of the packet flow is within the first parameter range, according to the predetermined mapping relationship between the characteristic parameter range and the packet priority , The message priority corresponding to the first parameter range is determined as the priority of the first message, and the mapping relationship between the characteristic parameter range and the message priority includes: the first parameter range and the The mapping relationship of the message priority corresponding to the first parameter range.
  • the third determining unit 404 is specifically configured to: obtain the value of the first parameter, and the value of the first parameter is used to indicate the priority of the second packet, and the second packet is the same as The first message belongs to the same message flow, and the second message is the previous message of the first message; the first message is determined according to the value of the first parameter and the characteristic parameter of the message flow. State the priority of the first message.
  • the device 400 further includes a first modification unit configured to modify the value of the first parameter, and the modified value of the first parameter is used to indicate the first parameter. The priority of the message.
  • the device 400 further includes a second modification unit configured to modify the value of the first field in the first message, and the modified value of the first field indicates the value of the first field. State the priority of the first message.
  • the device 400 further includes: a forwarding unit configured to forward the first message according to the priority of the first message.
  • the apparatus 400 further includes: a first sending unit configured to send the characteristic parameters of the packet stream to a display device for display.
  • the device 400 further includes: a second sending unit and an adjustment unit, where the second sending unit is used to send the characteristic parameters of the packet stream to the controller;
  • the controller obtains the classification result of the characteristic parameters of multiple message streams, adjusts the characteristic parameter range corresponding to the message flow, and the adjusted characteristic parameter range is used to determine the priority of the third message, so
  • the third packet is a packet obtained after the obtaining unit obtains the first packet, and the third packet belongs to the packet flow.
  • the device 400 further includes: a third sending unit, configured to send the actual delay of the target queue to the controller, so that the controller is in the actual delay of the target queue.
  • a third sending unit configured to send the actual delay of the target queue to the controller, so that the controller is in the actual delay of the target queue.
  • the target parameter is adjusted, and the adjusted target parameter is used to determine the priority of a third packet, and the third packet is obtained by the network device.
  • a message obtained after the first message, and the third message belongs to the message flow; wherein: the target queue is a queue in the network device for storing the message to be forwarded; the target parameter includes Any one or more of the following: corresponding to the characteristic parameter range of the packet flow and the preset time threshold corresponding to each characteristic parameter, the first characteristic parameter is any one of the characteristic parameters, and the first characteristic parameter is any one of the characteristic parameters.
  • a preset time threshold corresponding to a characteristic parameter is used to instruct the network device to clear the first characteristic parameter when the time interval between adjacent packets is greater than the preset time threshold.
  • the characteristic parameter of the packet flow includes any one or more of the following: the number of packets included in the burst of the packet flow, the maximum packet time interval of the packet flow
  • the message time interval is the time interval of adjacent messages, the rate fluctuation coefficient of the message flow, and the maximum instantaneous rate of the message flow.
  • the device 400 is a device corresponding to the message processing method corresponding to FIG. 2 described in the above embodiment
  • the specific implementation of each unit of the device 400 is based on the same idea as the above method embodiment. Therefore, regarding the For the specific implementation of each unit of the apparatus 400, reference may be made to the description of the packet processing method corresponding to FIG. 2 in the above method embodiment, and details are not described herein again.
  • FIG. 5 is a schematic structural diagram of a message processing device provided by an embodiment of this application.
  • the message processing device 500 includes: a processor 510, a communication interface 520, and a memory 530.
  • the number of processors 510 in the device 500 may be one or more.
  • One processor is taken as an example in FIG. 5.
  • the processor 510, the communication interface 520, and the memory 530 may be connected through a bus system or other methods.
  • the connection through the bus system 540 is taken as an example.
  • the processor 510 may be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and an NP.
  • the processor 510 may further include a hardware chip.
  • the above-mentioned hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof.
  • the above-mentioned PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.
  • the memory 530 may include a volatile memory (English: volatile memory), such as random-access memory (RAM); the memory 530 may also include a non-volatile memory (English: non-volatile memory), such as fast Flash memory (English: flash memory), hard disk drive (HDD) or solid-state drive (SSD); the memory 530 may also include a combination of the foregoing types of memory.
  • the memory 530 may store the aforementioned first flow table and the second flow table, for example.
  • the memory 530 stores an operating system and a program, an executable module or a data structure, or a subset of them, or an extended set of them, where the program may include various operation instructions for implementing various operations.
  • the operating system may include various system programs for implementing various basic services and processing hardware-based tasks.
  • the processor 510 may read the program in the memory 530 to implement the time synchronization method provided in the embodiment of the present application or the message processing method for time synchronization.
  • the bus system 540 may be a peripheral component interconnect standard (PCI) bus or an extended industry standard architecture (EISA) bus, etc.
  • PCI peripheral component interconnect standard
  • EISA extended industry standard architecture
  • the bus system 540 can be divided into an address bus, a data bus, a control bus, and so on. For ease of presentation, only one thick line is used to represent in FIG. 5, but it does not mean that there is only one bus or one type of bus.
  • the embodiments of the present application also provide a computer-readable storage medium, including instructions or computer programs, which when run on a computer, cause the computer to execute the message processing method provided in the above embodiments.
  • the embodiments of the present application also provide a computer program product containing instructions or computer programs, which when run on a computer, cause the computer to execute the message processing method provided in the above embodiments.
  • the device or device mentioned in this application may be a network device, such as a switch, a router, a server, or a part of a network device or server.
  • the device or device of the present application may also be a functional module deployed in the device or network.
  • the technical solution of the present application is currently applicable to the fields of wide area networks, campus networks and data centers, and the method can be applied to chips.
  • FIG. 6a is a schematic diagram of a network architecture provided by an embodiment of the application.
  • FIG. 6b is a schematic diagram of a network architecture of a data center network provided by an embodiment of the application.
  • the schematic diagram of the network architecture shown in Figure 6a is suitable for wide area networks and campus networks.
  • the client 607 can interact with the remote server 601, the data center server 602, and the near-end server 603 through the edge network device 606, the convergence layer network device 605, and the backbone layer network device 604.
  • the technical solutions provided by the embodiments of the present application can be applied to the edge network device 606, the convergence layer network device 605, and the backbone layer network device 604.
  • data traffic may include upstream traffic and downstream traffic.
  • the uplink traffic refers to the data traffic that is transferred from the edge network device 606 to the backbone layer network device 604 through the convergence layer network device 605.
  • Downlink traffic refers to the data traffic that is transferred from the backbone layer network device 604 to the edge network device 606 through the convergence layer network device 605.
  • the edge network device 606 applies the technical solution provided in the embodiments of this application, and the convergence layer network device 605 and the backbone layer network device 604 can correctly accept the message priority determined by the edge network device 606, then the convergence layer network device 605 and the backbone layer network device 604 may also not need to apply the solutions provided in the embodiments of the present application.
  • the edge network device 606 determines the priority of the first packet, it correspondingly modifies the value of the DSCP field of the first packet, and the convergence layer network device 605 and the backbone layer network device 604 can directly determine the first packet based on the value of the DSCP field.
  • the priority of a message For downstream traffic, if the backbone layer network device 604 applies the technical solution provided in the embodiments of this application, and the convergence layer network device 605 and the edge network device 606 can correctly accept the message priority determined by the backbone layer network device 604, Then, the convergence layer network device 605 and the edge network device 606 may not need to apply the solutions provided in the embodiments of the present application.
  • the backbone layer network device 604 determines the priority of the first packet, it correspondingly modifies the value of the DSCP field of the first packet.
  • the aggregation layer network device 605 and the edge network device 606 can directly determine the first packet based on the value of the DSCP field. The priority of a message.
  • the data center network may include a core layer 610, an aggregation layer 620, and an access layer 630.
  • the core layer 610, the aggregation layer 620, and the access layer 630 may all include multiple switches. Among them, one or more switches in the convergence layer 620 and one or more switches in the access layer may form a group, such as the group 640 and the group 650 shown in FIG. 6b.
  • the data exchange of the data center network occurs between the various switches of the data center network.
  • the technical solutions provided in the embodiments of the present application can be applied to the various switches shown in FIG. 6b.
  • the switch at the access layer 630 applies the technical solution provided in the embodiments of the present application, and the switch at the aggregation layer 620 and the switch at the core layer 610 can correctly accept the message priority determined by the switch at the access layer 630, Then, the switches at the convergence layer 620 and the switches at the core layer 610 may not need to apply the solutions provided in the embodiments of the present application.
  • FIG. 7a and FIG. 7b provide a schematic diagram of an exemplary application scenario for an embodiment of the application.
  • the flow identification and classification technology provided by this application can be embedded in other modules 702 before the flow management module 701.
  • the flow identification and classification technology provided in this application is applied to an independent flow classification module 703, which is located between the flow management module 701 and other modules 702.
  • the disclosed system, device, and method can be implemented in other ways.
  • the device embodiments described above are merely illustrative, for example, the division of units is only a logical business division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or integrated. To another system, or some features can be ignored, or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • business units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be realized in the form of hardware or software business unit.
  • the integrated unit is implemented in the form of a software business unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of the present application essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods in the various embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other media that can store program codes. .
  • the services described in the present invention can be implemented by hardware, software, firmware, or any combination thereof.
  • these services 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.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本申请实施例公开了一种报文处理方法。网络设备获取第一报文，确定第一报文所属的报文流，当连续获取的属于该报文流的报文数量大于或者等于第一阈值时，即当该报文流不是小流时，根据获取的第一报文确定报文流的特征参数，其中，连续获取的属于所述报文流的报文中，相邻报文的时间间隔小于或者等于第一时间阈值。根据报文流的特征参数确定报文流对传输时延的要求，根据报文流的特征参数确定第一报文的优先级。对于所属报文流不是小流的第一报文而言，不是像传统技术中那样，直接将第一报文的优先级确定为低优先级，而是结合第一报文所属的报文流的特征参数来确定第一报文的优先级，从而使得所确定的优先级能够符合第一报文对传输时延的要求。

Description

一种报文处理方法及装置
本申请要求于2019年12月20日提交中国专利局、申请号为201911329488.1、申请名称为“一种报文处理方法及装置”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
本申请涉及数据处理领域，尤其涉及一种报文处理方法及装置。
背景技术
网络设备可以对报文进行转发，由于网络设备转发的报文数量比较多，而网络设备同一时刻转发的报文数量是有限的。因此，对于待转发的报文，网络设备可以确定报文的优先级，并根据报文的优先级对报文进行转发。例如，网络设备可以根据优先级由高到低的顺序对报文进行转发。可以理解的是，对于优先级较高的报文，其对应的传输时延相对比较低，而对于优先级较低的报文，其对应的传输时延则可能比较高。
传统技术中确定报文优先级的方式，并不能满足一些报文对传输时延的要求。
发明内容
本申请实施例提供了一种报文处理方法，以改善报文的传输时延。
第一方面，本申请实施例提供了一种报文处理方法，该方法可以应用于路由器、交换机等网络设备。具体地，对于不是小流的报文流而言，其对传输时延的要求可以在报文流的特征参数上有所体现。因此，在本申请实施例中，网络设备获取第一报文，可以确定第一报文所属的报文流，当连续获取的属于该报文流的报文数量大于或者等于第一阈值时，即当该报文流不是小流时，可以进一步根据获取的第一报文确定报文流的特征参数。其中，连续获取的属于该报文流的报文中，相邻报文的时间间隔小于或者等于第一时间阈值。由于报文流的特征参数可以体现报文流对传输时延的要求，因此，确定报文流的特征参数之后，可以根据报文流的特征参数确定第一报文的优先级。例如，报文流的特征参数指示报文流要求传输时延比较低，则可以确定第一报文的优先级为较高优先级；又如，报文流的特征参数指示报文流允许传输时延比较高，则可以确定第一报文的优先级为低优先级，等等。由此可见，利用本申请实施例的方案，对于所属报文流不是小流的第一报文而言，不是像传统技术中那样，直接将第一报文的优先级确定为低优先级，而是结合第一报文所属的报文流的特征参数来确定第一报文的优先级，从而使得所确定的优先级能够符合第一报文对传输时延的要求。
在一种实现方式中，对于不是小流的报文流而言，根据报文流对传输时延的要求，可以将这些报文流划分成多个类别。在对报文流进行类别划分时，可以将对传输时延的要求等级基本相同的报文流划分成一个类别。这样一来，则可以根据报文流的类别确定报文的优先级，其中，一个类别可以对应一种优先级。可以理解的是，类别相同的报文流对传输时延的要求等级基本相同，而报文流的特征参数又可以体现报文流对传输时延的要求。因此，对于属于某一类别的一个或者多个报文流而言，该一个或者多个报文流的特征参数可以认为在一定的参数范围内。考虑到这种情况，在本申请实施例的又一种实现方式中，网络设备例如可以预先确定特征参数范围与报文优先级的映射关系，其中，特征参数范围可以体现报文流的类别。这样一来，网络设备确定报文流的特征参数之后，可以进一步确定 报文流的特征参数所属的第一参数范围，而后，网络设备根据预先确定的特征参数范围与报文优先级的映射关系，将与第一参数范围对应的报文优先级确定为第一报文的优先级。
在一种实现方式中，当相邻报文的时间间隔大于第一时间阈值时，网络设备可以对前述连续获取的属于该报文流的报文数量进行清零。另外，网络设备还可以根据特征参数对应的老化周期对报文流的特征参数进行清零。对于第一报文所属的报文流而言，若在获取第一报文之前，该连续获取的属于该报文流的报文数量已经清零，则获取到第一报文之后，该报文数量会从1重新计数，这将会影响确定该报文流是否是小流的结果，相应的，会影响确定第一报文的优先级的结果。另外，若在获取第一报文之前，该报文流的特征参数已经清零，则获取到第一报文之后，该报文流的特征参数会重新计算，这也会影响确定第一报文的优先级的结果。可以理解的是，第一报文的优先级若确定错误，则可能会导致第一报文所属的报文流发生乱序。所谓报文流乱序，指的是网络设备转发报文的顺序，与网络设备获取报文的顺序不一致。为了改善这种情况，在本申请实施例中，网络设备中可以存储第一参数，第一参数的值用于指示第二报文的优先级。相应的，网络设备可以结合该第一参数的值和报文流的特征参数确定第一报文的优先级。
在一种实现方式中，网络设备确定第一报文的优先级之后，还可以对前述第一参数的值进行修改，使得修改后的第一参数的值指示第一报文的优先级。对第一参数的值进行修改之后，若网络设备再次获取到属于该报文流的其它报文，则可以结合修改后的第一参数的值来确定前述其它报文的优先级，减少报文流乱序。
在一种实现方式中，第一报文中可以包括指示第一报文的优先级的第一字段。而网络设备获取的第一报文中包括的第一字段的值，可能与根据报文流的特征参数确定的第一报文的优先级不一致。因此，在本申请实施例中，网络设备确定的第一报文的优先级与第一报文中第一字段的值指示的报文优先级不吻合，则网络设备可以对该第一字段进行修改，使得修改后的第一字段的值指示根据报文流的特征参数所确定的第一报文的优先级。
在一种实现方式中，网络设备确定第一报文的优先级之后，网络设备可以根据第一报文的优先级转发该第一报文。
在一种实现方式中，网络设备确定该报文流的特征参数之后，除了可以根据该特征参数确定第一报文的优先级之外，还可以将该报文流的特征参数发送给显示设备，由显示设备对接收到的特征参数进行显示。这样一来，用户可以通过显示设备显示的内容对报文流进行分析。
在一种实现方式中，前述特征参数范围可以是控制器下发给网络设备的，为了使得所确定的报文的优先级更加准确和实时调整，在本申请实施例的一种实现方式中，控制器还可以根据报文流的特征参数对报文流的特征参数范围进行调整。具体地，网络设备确定报文流的特征参数之后，可以将该特征参数发送给控制器。可以理解的是，控制器可以接收到多个报文流的特征参数，因为网络设备每确定一个报文流的特征参数，都可以将该特征参数发送给控制器。控制器可以根据获取到的多个报文流的特征参数进行分类。而后网络设备可以根据该分类结果对前述特征参数范围进行调整。网络设备获得调整后的特征参数之后，若网络设备继续获取到属于该报文流的第三报文，则可以根据该调整后的特征参数范围确定第三报文的优先级。
在一种实现方式中，一方面，前述提及的特征参数范围和预设时间周期均可能会影响确定第一报文的优先级的结果，从而导致第一报文的转发时延不能满足第一报文对传输时延的要求。另一方面，若网络设备确定报文优先级的结果比较准确，则网络设备中各个队列的实际时延小于或者等于各个队列对应的时延上限。若网络设备中存在实际时延大于其对应的时延上限的队列，则表示前述特征参数范围或者预设时间周期的值不合理。因此，在本申请实施例中，网络设备还可以将目标队列的实际时延发送给控制器，由控制器根据目标队列的时延和目标队列的时延上限对目标参数进行调整，从而使得根据调整后的目标参数确定的报文优先级，可以使得目标队列的实际时延小于目标队列的时延上限。具体地，控制器可以在目标队列的实际时延大于时延上限时，对目标参数进行调整。此处提及的目标参数可以包括报文流的特征参数范围和预设时间周期中的任意一个或者多个。控制器对目标参数进行调整之后，可以将调整之后的目标参数发送给网络设备，这样一来，若网络设备继续获取到属于该报文流的第三报文，则可以根据该调整后的目标参数确定第三报文的优先级。
在一种实现方式中，报文流的特征参数可以是能够体现该报文流对传输时延的要求的一个或者多个参数。一些允许传输时延比较高的业务例如文件下载业务，其会产生比较大的数据流量，为了减少处理这些数据流量所需的中央处理器(central processing unit，CPU)开销，处理该业务的服务器往往会阶段性批量发送大量的报文，这就导致该业务对应的报文流的突发较大，而且最大报文时间间隔也较大。此处提及的报文时间间隔指的是相邻报文的时间间隔。而对于一些要求传输时延比较小的业务例如视频会议业务，其产生的流量相对于视频下载等业务而言较小，所以对应的突发较小。另外，由于这些业务对实时性要求比较高，因此，处理该业务的服务器往往会实时地、连续不断地发送报文，这就导致该业务对应的报文流的最大报文时间间隔也较小。报文流的突发包括的报文数量以及最大报文时间间隔均可以体现报文流对应业务对传输时延的要求，因此，报文流的特征参数可以包括该报文流的突发包括的报文数量。报文流的特征参数还可以包括最大报文时间间隔。另外，通常要求传输时延比较低的报文流的速率一般会比较平稳，而允许传输时延比较高的报文流的速率波动比较大。因此，报文流的特征参数还可以包括报文流的速率波动系数。另外，要求传输时延比较低的报文流的最大瞬时速率一般不会很大，而允许传输时延比较高的报文流的最大瞬时速率可能非常大。报文流对传输时延的要求还可以结合报文流的瞬时速率来确定。因此，报文流的特征参数还可以包括报文流的最大瞬时速率。
第二方面，本申请实施例提供了一种报文处理装置，所述装置包括：获取单元、第一确定单元、第二确定单元和第三确定单元。获取单元用于获取第一报文；第一确定单元用于确定所述第一报文所属的报文流；第二确定单元用于当连续获取的属于所述报文流的报文数量大于或者等于第一阈值时，根据所述第一报文确定所述报文流的特征参数，所述连续获取的属于所述报文流的报文中，相邻报文的时间间隔小于或者等于第一时间阈值；第三确定单元用于根据所述报文流的特征参数确定所述第一报文的优先级。
在一种实现方式中，所述第三确定单元具体用于：当所述报文流的特征参数在第一参数范围内时，根据预先确定的特征参数范围与报文优先级的映射关系，将所述第一参数范围对应的报文优先级确定为所述第一报文的优先级，所述特征参数范围与报文优先级的映 射关系包括：所述第一参数范围和所述第一参数范围对应的报文优先级的映射关系。
在一种实现方式中，所述第三确定单元具体用于:获取第一参数的值，所述第一参数的值用于指示第二报文的优先级，所述第二报文与所述第一报文属于同一报文流，所述第二报文为所述第一报文的前一条报文；根据所述第一参数的值和所述报文流的特征参数确定所述第一报文的优先级。
在一种实现方式中，所述装置还包括第一修改单元，第一修改单元用于对所述第一参数的值进行修改，修改后的第一参数的值用于指示所述第一报文的优先级。
在一种实现方式中，所述装置还包括第二修改单元，第二修改单元用于修改所述第一报文中第一字段的值，修改后的所述第一字段的值指示所述第一报文的优先级。
第二修改单元，所述装置还包括转发单元，转发单元用于根据所述第一报文的优先级转发所述第一报文。
在一种实现方式中，所述装置还包括第一发送单元，第一发送单元用于将所述报文流的特征参数发送给显示设备进行显示。
在一种实现方式中，所述装置还包括第二发送单元和调整单元。第二发送单元用于将所述报文流的特征参数发送给控制器；调整单元用于根据从所述控制器获取的、对多个报文流的特征参数的分类结果，调整对应所述报文流的特征参数范围，调整后的对应所述报文流的特征参数范围用于确定第三报文的优先级，所述第三报文为所述获取单元获取所述第一报文之后获取的报文，所述第三报文属于所述报文流。
在一种实现方式中，所述装置还包括第三发送单元。第三发送单元用于将目标队列的实际时延发送给控制器，以使得所述控制器在所述目标队列的实际时延大于所述目标队列的时延上限时对目标参数进行调整，调整后的所述目标参数用于确定第三报文的优先级，所述第三报文为所述网络设备获取所述第一报文之后获取的报文，所述第三报文属于所述报文流；其中：所述目标队列为所述网络设备中用于存储待转发报文的队列；所述目标参数包括以下任意一项或者多项：对应所述报文流的特征参数范围和各个特征参数分别对应的预设时间阈值，第一特征参数为所述各个特征参数中的任意一个特征参数，所述第一特征参数对应的预设时间阈值，用于指示所述网络设备在相邻报文的时间间隔大于所述预设时间阈值时，对所述第一特征参数进行清零。
在一种实现方式中，所述报文流的特征参数包括以下任意一项或者多项：所述报文流的突发包括的报文数量、所述报文流的最大报文时间间隔，所述报文时间间隔为相邻报文的时间间隔、所述报文流的速率波动系数、以及所述报文流的最大瞬时速率。
第三方面，本申请实施例提供了一种设备。所述设备包括处理器和存储器。所述存储器用于存储指令或计算机程序。所述处理器用于执行所述存储器中的所述指令或计算机程序，执行以上第一方面任意一项所述的方法。
第四方面，本申请实施例提供了一种计算机可读存储介质，包括指令或计算机程序，当其在计算机上运行时，使得计算机执行以上第一方面任意一项所述的方法。
第五方面，本申请实施例提供了一种包含指令或计算机程序的计算机程序产品，当其在计算机上运行时，使得计算机执行以上第一方面任意一项所述的方法。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请中记载的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为一个示例性应用场景示意图；
图2为本申请实施例提供的一种报文处理方法的流程示意图；
图3为本申请实施例提供的一种确定报文流的突发的方法的示意图；
图4为本申请实施例提供的一种报文处理装置的结构示意图；
图5为本申请实施例提供的一种报文处理设备的结构示意图；
图6a为本申请实施例提供的一种网络架构示意图；
图6b为本申请实施例提供的一种数据中心网络的网络架构示意图；
图7a为本申请实施例提供一种示例性应用场景示意图；
图7b为本申请实施例提供一种示例性应用场景示意图。
具体实施方式
本申请实施例提供了一种报文处理方法，用于改善报文的传输时延。
为方便理解，首先对本申请实施例的可能的应用场景进行介绍。参见图1，该图为一个示例性应用场景示意图。在介绍图1所示的场景之前，首先对以下实施例涉及的“报文流”的概念进行介绍。
报文在传输的过程中，可以携带与报文传输相关的信息，该信息可以为二元组、五元组或者七元组。其中，二元组包括源互联网协议(Internet Protocol，IP)地址和目的IP地址。五元组包括源IP地址、源端口号、目的IP地址、目的端口号和传输层协议号。七元组包括源媒体访问控制(media access control，MAC)地址、目的MAC地址、源IP地址、目的IP地址、协议号、源端口号和目的端口号。若在两个节点之间连续传输的多条报文，其携带的上述信息均相同，则该连续报文也可以被称为报文流。因此，可以通过报文携带的信息例如二元组、五元组、七元组或者可以用于区分报文流的其它信息来确定该报文所属的报文流。
接下来对图1所示的场景进行介绍。
在图1所示的场景中，设备101可以从设备102处获取报文，并将该报文转发至设备103。本申请实施例不具体限定设备101、设备102以及设备103的具体形态。作为一种示例，设备101例如可以为路由器或者交换机等网络设备。设备102可以为路由器、交换机等网络设备，也可以为服务器或者终端设备等设备。设备103可以为路由器或者交换机等网络设备，也可以为服务器或者终端设备等设备。设备101获取的报文数量比较多，而设备101的报文转发能力有限，即设备101在同一时刻转发的报文数量是有限的，对于设备101待转发的报文，设备101可以确定各个报文的优先级，并根据各个报文的优先级对报文进行转发。
需要说明的是，图1只是为了方便理解本申请实施例的应用场景而示出，其并不构成对本申请实施例的限定。
设备101获取的报文可以对应多个业务，例如视频点播业务或者语音业务等等。不同业务对流完成时间(flow completion time，FCT)的要求不同，例如，语音通话业务要求报文流完成时间比较短，而视频点播业务则允许流完成时间比较长。而报文流的流完成时间可以通过该报文流包括的报文的传输时延来体现。不同业务的报文对传输时延的要求不同，例如语音通话业务要求传输时延比较低，而视频点播业务则允许传输时延比较高。而且，要求传输时延比较低的业务对应的报文流一般情况下为小流。所谓小流，指的是速率比较低的报文流、或者包括报文数量比较少的报文流。因此，在传统技术中，确定报文优先级的方法比较简单，即：若报文所属的报文流为小流，则该报文的优先级为高优先级，若报文所属的报文流不是小流，则该报文的优先级为低优先级。
然而，对于一些业务而言，虽然其对应的报文流不是小流，但是该业务也要求传输时延比较低。例如，对于视频会议业务以及虚拟现实(virtual reality，VR)等业务而言，其对应的报文流虽然不是小流，但是这类业务也要求传输时延比较低。而采用传统技术中的确定报文优先级的方案，会将这类业务的报文的优先级确定为低优先级，从而使得所确定的优先级并不能满足报文对传输时延的要求，进一步地，会影响业务的服务质量(quality of service，QOS)。
鉴于此，本申请实施例提供了一种报文处理方法，该方法用于确定报文的优先级，所确定的报文的优先级可以满足报文对传输时延的要求。具体地，利用本申请实施例的方案，可以在当待转发的第一报文所属的报文流不是小流时，所确定的第一报文的优先级可以满足第一报文对传输时延的要求。以下结合附图对本申请实施例提供的报文处理方法进行介绍。
参见图2，该图为本申请实施例提供的一种报文处理方法的流程示意图。图2所示的报文处理方法，例如可以通过如下S101-S103实现。该实施例以网络设备为执行主体为例。
S101：网络设备获取第一报文，确定所述第一报文所属的报文流。
在本申请实施例中，网络设备可以为对第一报文进行转发的中间节点。网络设备可以从其它设备处接收第一报文，网络设备获取到第一报文之后，可以将第一报文转发给其它设备。本申请实施例中提及的网络设备例如可以为图1所示的设备101。
如上所述，不同业务对流完成时间的要求不同，相应的，不同业务的报文对传输时延的要求不同。因此，在本申请实施例中，网络设备获取到第一报文之后，可以进一步确定第一报文所属的报文流，并根据第一报文所属的报文流来确定第一报文的优先级。具体地，网络设备可以根据第一报文的二元组、五元组或者七元组来确定第一报文所属的报文流。
S102：当连续获取的属于该报文流的报文数量大于或者等于第一阈值时，网络设备根据第一报文确定该报文流的特征参数，该连续获取的属于所述报文流的报文中，相邻报文的时间间隔小于或者等于第一时间阈值。
S103：网络设备根据该报文流的特征参数确定第一报文的优先级。
在本申请实施例中，若第一报文所属的报文流中相邻报文之间的时间间隔比较小，且报文数量比较多时，可以确定第一报文所属的报文流不是小流。具体地，网络设备确定第一报文所属的报文流之后，可以进一步确定连续获取的属于该报文流的报文数量是否小于或者等于第一阈值，以确定第一报文所属的报文流是否为小流，其中，连续获取的属于该 报文流的报文中，相邻报文的时间间隔小于或者等于第一时间阈值。关于第一时间阈值的具体取值可以根据实际情况确定，该第一时间阈值例如可以为500毫秒。
在本申请实施例中，若第一报文所属的报文流包括的报文数量小于第一阈值，则表示第一报文所属的报文流是小流，对于这种情况，可以直接确定第一报文的优先级为高优先级。
若第一报文所属的报文流包括的报文数量大于或者等于第一阈值，则表示第一报文所属的报文流不是小流。在本申请实施例中，若该报文流不是小流，则进一步根据报文流对传输时延的要求确定第一报文的优先级。具体地，报文流的特征参数可以体现报文流对传输时延的要求，因此，可以进一步确定该报文流的特征参数，从而进一步确定第一报文的优先级。
关于报文流的特征参数，本申请实施例不做具体限定，报文流的特征参数可以是能够体现该报文流对传输时延的要求的一个或者多个参数。一些允许传输时延比较高的业务例如文件下载业务，其会产生比较大的数据流量，为了减少处理这些数据流量所需的中央处理器(central processing unit，CPU)开销，处理该业务的服务器往往会阶段性批量发送大量的报文，这就导致该业务对应的报文流的突发(burst)较大，而且最大报文时间间隔也较大。此处提及的报文时间间隔指的是相邻报文的时间间隔。而对于一些要求传输时延比较小的业务例如视频会议业务，其产生的流量相对于视频下载等业务而言较小，所以对应的突发较小。另外，由于这些业务对实时性要求比较高，因此，处理该业务的服务器往往会实时地、连续不断地发送报文，这就导致该业务对应的报文流的最大报文时间间隔也较小。
关于报文流的突发，需要说明的是，一个报文流可以划分成一个或者多个突发。若相邻报文之间的时间间隔小于或者等于T，则这两个报文属于同一个突发。此处提及的T可以理解为与突发包括的报文数量对应的老化周期。如下图3所示，图3为本申请实施例提供的一种确定报文流的突发的方法的示意图。如图3所示，横轴表示获取报文的时间，纵轴表示突发包括的报文数量，即burst size。横轴下方的小方块表示报文，按照从左到右的顺序，前四个报文中相邻报文的时间间隔都小于T，而第四个报文与第五个报文的时间间隔大于T，因此，可以确定突发包括的报文数量为4，即counter＝4。相应的，从第五个报文至第10个报文，相邻报文的时间间隔都小于T，因此，可以确定counter＝6。上文提及的突发较大，指的是突发包括的报文数量比较多，突发较小指的是突发包括的报文数量比较少。
通过上述描述可知，报文流的突发包括的报文数量以及最大报文时间间隔均可以体现报文流对应业务对传输时延的要求，因此，在本申请实施例的一种实现方式中，报文流的特征参数可以包括该报文流的突发包括的报文数量。在本申请实施例的又一种实现方式中，报文流的特征参数还可以包括最大报文时间间隔。
关于最大报文时间间隔的确定方式，需要说明的是，假设g表示一条报文流的最大报文时间间隔，则当网络设备获取到该报文流的一个新报文时，若该新报文的时间与上一条报文的时间间隔大于t1，则将g清零，否则将g的值更新为g＝max{g,△t}，其中△t为该新报文与前一个报文的时间间隔。此处提及的t1可以理解为与最大报文时间间隔对应的老 化周期。
一般要求传输时延比较低的报文流的速率一般会比较平稳，而允许传输时延比较高的报文流的速率波动比较大。因此，在本申请实施例的又一种实现方式中，报文流的特征参数还可以包括报文流的速率波动系数，该速率波动系数例如可以为报文流的速率的方差，又如可以为报文流的速率的标准差，再如可以为报文流的标准差与报文流平均速率的比值。另外，要求传输时延比较低的报文流的最大瞬时速率一般不会很大，而允许传输时延比较高的报文流的最大瞬时速率可能非常大。报文流对传输时延的要求还可以结合报文流的瞬时速率来确定。因此，在本申请实施例的又一种实现方式中，报文流的特征参数还可以包括报文流的最大瞬时速率。
其中，报文流的速率可以通过如下方式确定：确定在时间段T1内报文流所包括的报文的字节数，利用该字节数除以T1即可得到报文流的速率，可以理解的是，若T1的取值很小，则可以计算得到报文流的瞬时速率，若T1比较大，则可以计算得到报文流的平均速率。计算得到报文流的速率之后，则可以进一步得到报文流的最大瞬时速率和报文流的速率波动系数。
在一些实施例中，对于S103，网络设备中可以预先存储特征参数与报文优先级的映射关系。网络设备确定第一报文所属的报文流的特征参数，根据该预先存储的映射关系确定第一报文的优先级。该映射关系中至少包括S102中确定的特征参数以及S103中所确定的第一报文的优先级之间的映射关系。
另外，对于不是小流的报文流而言，根据报文流对传输时延的要求，可以将这些报文流划分成多个类别。在对报文流进行类别划分时，可以将对传输时延的要求等级基本相同的报文流划分成一个类别。这样一来，则可以根据报文流的类别确定报文的优先级，其中，一个类别可以对应一种优先级。类别相同的报文流对传输时延的要求等级基本相同，而报文流的特征参数又可以体现报文流对传输时延的要求。因此，对于属于某一类别的一个或者多个报文流而言，该一个或者多个报文流的特征参数可以认为在一定的参数范围内。考虑到这种情况，在本申请实施例的又一种实现方式中，网络设备例如可以预先确定特征参数范围与报文优先级的映射关系，其中，特征参数范围可以体现报文流的类别。这样一来，网络设备确定报文流的特征参数之后，可以进一步确定报文流的特征参数所属的第一参数范围，而后，网络设备根据预先确定的特征参数范围与报文优先级的映射关系，将与第一参数范围对应的报文优先级确定为第一报文的优先级。
关于报文流的类别，现举例说明，报文流的类别例如可以为时延敏感类，表示该报文流对应的业务对时延比较敏感，即该报文流要求传输时延比较低；报文流的类别又如可以为时延不敏感类，表示该报文流对应的业务对时延不敏感，即该报文流允许传输时延比较高。当然，根据报文流对应的业务对传输时延的敏感程度，时延敏感类还可以细分成时延高敏感类、时延中敏感类以及时延低敏感类等等，此处不再一一列举说明。
关于报文流的类别和特征参数范围之间的对应关系，可以结合下表1进行理解。关于特征参数范围和报文优先级的映射关系，可以结合下表2进行理解。
需要说明的是，表1和表2只是为了方便理解而示出，其并不构成对本申请实施例的限定。报文流的特征参数可以既包括突发包括的报文数量、又包括最大报文时间间隔，当 然，还可以包括前文提及的速率波动系数以及最大瞬时速率。另外，报文流类别除了可以是表1示出的时延高敏感类、时延中敏感类和时延不敏感类之外，还可以包括其它类别，例如，还可以包括时延低敏感类等等，此处不再一一列举说明。特征参数范围也不限于表2所示，报文优先级除了可以是表2所示的较高、中、低之外，还可以是高、中以及较低等等。另外，表1只是为了说明特征参数范围所表示的物理含义而示出，实际上网络设备可以无需根据表1所示的映射关系确定报文流的类别。网络设备可以直接根据表2所示的映射关系确定第一报文的优先级。
表1
特征参数范围 类别
突发包括的报文数量小于或者等于8 时延高敏感类
突发包括的报文数量大于8且小于或者等于39 时延中敏感类
突发包括的报文数量大于39 时延不敏感类
最大报文时间间隔大于或者等于500微秒 时延不敏感类
最大报文时间间隔小于500微秒 时延中敏感类
表2
特征参数范围 报文优先级
突发包括的报文数量小于或者等于8 较高
突发包括的报文数量大于8且小于或者等于39 中
突发包括的报文数量大于39 低
关于确定连续获取的属于前述报文流的报文数量是否大于或者等于第一阈值的具体实现方式，需要说明的是，在本申请实施例的一种实现中，网络设备中可以具备用于存储前述报文数量的第一流表(flow table)。在一些实施例中，第一报文所属的报文流可以包括多条报文，而第一报文可以是该报文流的第一条报文，也可以不是该报文流的第一条报文，当第一报文是该报文流的第一条报文时，网络设备获取到第一报文之后，可以确定该报文数量为1，并将该报文数量存储到第一流表中。若第一报文不是该报文流的第一条报文，则前述第一流表中可能已经存储了网络设备获取第一报文之前对应的报文数量，假设第一流表中存储的报文数量为N，则获取到第一报文之后，网络设备可以确定获取第一报文与获取第二报文的时间间隔是否大于第一时间阈值，若大于，网络设备可以将第一流表中存储的报文数量清零，若小于或者等于，网络设备可以确定该报文数量为N+1。此处提及的第一时间阈值，也可以被称为报文数量的老化周期。此处提及的第二报文为第一报文的前一条报文，第二报文和第一报文属于同一报文流。
需要说明的是，在本申请实施例中，网络设备将报文数量存储到第一流表中时，例如可以首先对第一报文的二元组、五元组或者七元组进行哈希运算，得到对应的报文数量存储地址，而后获取该地址中存储的报文数量的值，并在该值的基础上加1，或者将该值清零。具体地，确定第一流表中存储的报文数量在具体实现时，可以利用经典的略图(sketch)技术来确定。例如可以利用sketch技术中的计数最小略图(count min sketch，CM sketch) 技术来确定。
另外，在本申请实施例中，对于不是小流的报文流而言，网络设备中可以具备用于存储报文流的特征参数的第二流表。具体地，第二流表中可以存储该报文流的标识和该报文流的特征参数。本申请实施例中提及的报文流的标识，例如可以为报文流的二元组、五元组或者七元组。因此，在本申请实施例的一种实现方式中，网络设备确定第一报文所属的报文流之后，可以进一步确定第二流表中是否存在与前述二元组、五元组或者七元组对应的特征参数。若存在，则网络设备也可以直接确定该报文流不是小流。
在本申请实施例中，网络设备获取到第一报文之后，可以首先确定第二流表中是否存在第一报文所属的报文流的特征参数，若存在，则网络设备可以直接确定第一报文所属的报文流不是小流。对于这种情况，在获取第一报文之前，第一报文所属的报文流已经确定不是小流。前述连续获取的属于该报文流的报文为在获取第一报文之前获取的报文。若第二流表中不存在第一报文所属的报文流的特征参数，则网络设备可以进一步查询第一流表，确定获取第一报文之前该报文流包括的报文数量N，而后，若网络设备获取第一报文与获取第二报文的时间间隔小于或者等于第一时间阈值，则网络设备可以在数量N的基础上加1，并进一步判断N+1是否大于或者等于第一阈值，若N+1大于或者等于第一阈值，则网络设备可以确定第一报文所属的报文流不是小流。若N+1小于第一阈值，则网络设备可以确定第一报文所属的报文流是小流。若网络设备获取第一报文与获取第二报文的时间间隔大于第一时间阈值，则网络设备将第一流表中的数量N清零，并确定第一报文所属的报文流为小流。对于这种情况，在获取第一报文之前，第一报文所属的报文流并未被确定不是小流，前述连续获取的属于该报文流的报文包括第一报文。
存储报文流的特征参数会占用网络设备的部分存储空间，因此，网络设备可以对报文流的特征参数进行清零。具体地，若报文流的特征参数包括多个特征参数，则各个特征参数均可以对应一个预设时间间隔。为方便描述，将各个特征参数中的任意一个特征参数称为第一特征参数，则第一特征参数对应的预设时间间隔，用于指示对报文流的第一特征参数进行清零，具体地，当该报文流的相邻报文间的时间间隔大于预设时间间隔时，对报文流的第一特征参数进行清零。在本申请实施例中，预设时间间隔可以是网络设备通过人机交互接口获取的，也可以是控制器发送给网络设备的，本申请实施例不做具体限定。本申请实施例不具体限定预设时间阈值的具体取值，预设时间阈值的具体取值可以根据实际情况确定，例如预设时间阈值的具体取值可以为200微秒。此处提及的第一特征参数对应的预设时间间隔，也可以被称作第一特征参数的老化周期。
在本申请实施例中，前述第一阈值以及特征参数范围可以是控制器下发给网络设备的，当然也可以是网络设备通过人机交互接口获取的，本申请实施例不做具体限定。
通过以上描述可知，利用本申请实施例的方案，对于所属报文流不是小流的第一报文，不是像传统技术中那样，直接将第一报文的优先级确定为低优先级，而是结合第一报文所属的报文流的特征参数来确定第一报文的优先级，从而使得所确定的优先级能够符合第一报文对传输时延的要求。
在一些实施例中，第一报文中可以包括指示第一报文的优先级的第一字段。而网络设备获取的第一报文中包括的第一字段的值，可能与S103中确定的第一报文的优先级不一 致。例如，向网络设备发送第一报文的设备确定报文优先级的方法与传统技术相同，故而确定的第一报文的优先级为低优先级。因此，在本申请实施例中，网络设备确定第一报文的优先级之后，若所确定的第一报文的优先级与第一报文中第一字段的值指示的报文优先级不吻合，则网络设备可以对该第一字段进行修改，使得修改后的第一字段的值指示S103所确定的第一报文的优先级。本申请实施例不具体限定该第一字段，该第一字段例如可以为区分服务编码点(differentiated services code point，DSCP)字段。
另外，确定第一报文的优先级之后，网络设备可以根据确定的优先级转发该第一报文。具体地，网络设备可以将第一报文加入与第一报文的优先级对应的队列中，从而实现根据第一报文的优先级转发第一报文的目的。
在一些实施例中，若第一报文中的第一字段的值指示S103所确定的第一报文的优先级，则网络设备转发第一报文之后，若接收到第一报文的网络设备能够正确解析第一字段的值，接收到该第一报文的网络设备则可以根据该第一字段的值确定第一报文的优先级，并进一步基于该第一报文的优先级对第一报文进行转发，从而使得在对第一报文的转发过程中，其优先级均可以符合第一报文的报文流对传输时延的要求。
如前文，在本申请实施例中，当相邻报文的时间间隔大于第一时间阈值时，网络设备可以对前述连续获取的属于该报文流的报文数量进行清零。另外，网络设备还可以根据特征参数对应的老化周期对报文流的特征参数进行清零。对于第一报文所属的报文流而言，若在获取第一报文之前，前述报文数量已经清零，则获取到第一报文之后，该报文数量会从1重新计数，这将会影响确定该报文流是否是小流的结果，相应的，会影响确定第一报文的优先级的结果。另外，若在获取第一报文之前，该报文流的特征参数已经清零，则获取到第一报文之后，该报文流的特征参数会重新计算，这也会影响确定第一报文的优先级的结果。
第一报文的优先级若确定错误，则可能会导致第一报文所属的报文流发生乱序。所谓报文流乱序，指的是网络设备转发报文的顺序，与网络设备获取报文的顺序不一致。
关于由于第一报文的优先级确定错误导致的报文流乱序的问题，现举例说明。第一报文的前一条报文为第二报文，第二报文与第一报文属于同一报文流。在获取第一报文之前，第一报文所属的报文流的特征参数已经清零。假设在获取第一报文之前，该报文流对应的类别为时延不敏感类，相应的，第二报文的优先级被确定为低优先级。获取第一报文之后，由于第一报文所属的报文流的特征参数已经清零，则第一报文所属的报文流的特征参数会重新计算，若根据重新计算的特征参数确定的第一报文的优先级为较高优先级。这就导致网络设备转发第一报文的时间比转发第二报文的之间要早，从而导致报文流乱序。
为了改善这种情况，在本申请实施例中，网络设备中可以存储第一参数，第一参数的值用于指示第二报文的优先级。相应的，网络设备可以结合该第一参数的值和报文流的特征参数确定第一报文的优先级。具体地，可以分别根据第一参数的值来确定第一优先级，以及根据报文流的特征参数确定第二优先级，并将第一优先级和第二优先级中的较低优先级，确定为第一报文的优先级。采用这种方式可以保证第一报文的优先级不比第二报文的优先级高，从而可以避免报文流乱序。例如：第一参数的值指示第二报文的优先级为低优先级，即获取第一报文之前，该报文流的类别为时延不敏感类，则根据第一参数的值确定 的第一优先级为低优先级。而根据报文流的特征参数确定的第二优先级为较高优先级，故而可以确定第一报文的优先级为低优先级。因为第二优先级之所以是较高优先级，可能是因为在获取第一报文之前，该报文流的特征参数已经清零而导致的。又如：第一参数的值指示第二报文的优先级为较高优先级，即获取第一报文之前，该报文流的类别为时延高敏感类，根据第一参数的值确定的第一优先级为较高优先级。而根据报文流的特征参数确定的第二优先级为低优先级，故而可以确定第一报文的优先级为低优先级。因为根据获取第一报文之前的特征参数，还不能确定该报文流的类别为时延不敏感类，获取第一报文之后，根据第一报文计算的特征参数则可以确定该报文流的类别为时延不敏感类。比如，获取第一报文之前突发包括的报文数量为39，故而确定该报文流的类别为时延敏感类，而获取第一报文之后，突发包括的报文数量为40，故而可以确定该报文流的类别为时延不敏感类。
相应的，网络设备还可以对前述第一参数的值进行修改，使得修改后的第一参数的值指示第一报文的优先级。可以理解的是，对第一参数的值进行修改之后，若网络设备再次获取到属于该报文流的其它报文，则可以结合修改后的第一参数的值来确定前述其它报文的优先级，避免报文流乱序。
需要说明的是，在本申请实施例中，第一参数的值例如可以存储在前文提及的第二流表中，具体地，第二流表中可以存储第一报文所属的报文流的标识与该第一参数的值的映射关系。其中，报文流的标识可以为该报文流的二元组、五元组或者七元组。
当然，在本申请实施例中，为了节省存储该第一参数的空间，网络设备也可以按照一定周期例如第一周期对第一参数的值进行清零，相应的释放用于存储第一参数的值的内存空间。对第一参数进行清零时，还可以相应的删除报文流的标识。该第二周期的具体取值本申请实施例不做具体限定，该第一周期的值例如可以为保证报文流不会发生乱序的值，例如，第一周期的值可以为2毫秒。
在本申请实施例的一种实现方式中，网络设备确定该报文流的特征参数之后，除了可以根据该特征参数确定第一报文的优先级之外，还可以将该报文流的特征参数发送给显示设备，由显示设备对接收到的特征参数进行显示。具体地，显示设备例如可以以条形图、折线图等图形形式显示前述特征参数。可以理解的是，用户可以通过显示设备显示的内容对报文流进行分析。特别地，当报文流的特征参数包括突发包括的报文数量时，网络设备可以获取该报文流的各个突发包括的报文数量，并将各个突发分别包括的报文数量发送给显示设备进行显示。另外，当报文流的特征参数包括突发包括的报文数量时，显示设备所显示的内容除了可以是突发包括的报文数量之外，还可以是突发包括的报文总字节数。其中，突发包括的报文总字节数可以根据每条报文的平均字节数乘以突发包括的报文数量得到。这样一来，用户可以通过显示设备显示的内容确定报文的突发分布情况。本申请实施例不具体限定该显示设备，该显示设备可以为具备显示功能的设备，例如，该显示设备可以为终端设备。
如前文，前述特征参数范围可以是控制器下发给网络设备的，对于这种情况，为了使得所确定的报文的优先级更加准确，在本申请实施例的一种实现方式中，控制器还可以根据报文流的特征参数对该报文流的特征参数范围进行调整。具体地，网络设备确定报文流的特征参数之后，可以将该特征参数发送给控制器。可以理解的是，控制器可以接收到多 个报文流的特征参数，因为网络设备每确定一个报文流的特征参数，都可以将该特征参数发送给控制器。控制器可以根据获取到的多个报文流的特征参数进行分类。而后网络设备可以根据该分类结果对前述特征参数范围进行调整。本申请实施例不具体限定控制器对特征参数进行分类所使用的分类算法，该分类算法例如可以为经典的支持向量机(support vector machine，SVM)算法。现以报文流的特征参数包括突发包括的报文数量为例进行说明，控制器对多个报文流的突发包括的报文数量分为三类，第一类对应的突发包括的报文数量均小于或者等于9，第二类对应的突发包括的报文数量大于9且小于或者等于40，第三类对应的突发包括的报文数量大于40。则调整后特征参数范围，可以参考表3所示的特征参数范围和报文流的类别的映射关系进行理解。
表3
特征参数范围 类别
突发包括的报文数量小于或者等于9 时延高敏感类
突发包括的报文数量大于9且小于或者等于40 时延中敏感类
突发包括的报文数量大于40 时延不敏感类
可以理解的是，网络设备根据分类结果调整特征参数范围，可以有两种实现方式。作为一种实例，可以由控制器对特征参数范围进行调整，控制器对特征参数范围进行调整之后，可以将调整之后的特征参数范围发送给网络设备。作为又一种示例，控制器可以将前述分类结果发送给网络设备，网络设备根据该分类结果对特征参数进行调整。网络设备获得调整后的特征参数之后，若网络设备继续获取到属于该报文流的第三报文，则可以根据该调整后的特征参数范围确定第三报文的优先级。
在一些场景中，前述提及的特征参数范围和预设时间周期均可能会影响确定第一报文的优先级的结果，从而导致第一报文的转发时延不能满足第一报文对传输时延的要求。另一方面，若网络设备确定报文优先级的结果比较准确，则网络设备中各个队列的实际时延小于或者等于各个队列对应的时延上限。若网络设备中存在实际时延大于其对应的时延上限的队列，则表示前述特征参数范围或者预设时间周期的值不合理。因此，在本申请实施例中，网络设备还可以将目标队列的实际时延发送给控制器，由控制器根据目标队列的时延和目标队列的时延上限对目标参数进行调整，从而使得根据调整后的目标参数确定的报文优先级，可以使得目标队列的实际时延小于目标队列的时延上限。具体地，控制器可以在目标队列的实际时延大于时延上限时，对目标参数进行调整。此处提及的目标参数可以包括前述特征参数范围和预设时间周期中的任意一个或者多个。
关于目标队列，需要说明的是，目标队列可以包括网络设备中的一个或者多个队列，例如目标队列可以包括前文提及的队列a、队列b以及队列c中的任意一个或者多个队列。目标队列的时延上限可以是预先确定的，该时延上限可以是一个经验值，也可以是控制器通过对网络设备中目标队列的历史时延数据进行统计得到的，本申请实施例不做具体限定。目标队列的实际时延可以是网络设备计算得到的，网络设备可以将目标队列的瞬时队列时延确定为目标队列的实际时延。具体地，网络设备可以获取目标队列的报文深度，即目标队列包括的报文字节总数，并获取目标队列的瞬时速率，利用目标队列的报文深度除以目 标队列的瞬时速率即可得到目标队列的瞬时队列时延。
在本申请实施例中，控制器对目标参数进行调整之后，可以将调整之后的目标参数发送给网络设备，这样一来，若网络设备继续获取到属于该报文流的第三报文，则可以根据该调整后的目标参数确定第三报文的优先级。
关于控制器调整目标参数的方式，现举例说明。目标队列为前文提及的队列b，队列b对应的报文流的类别为时延敏感类。目标队列的实际时延为2.3毫秒，目标队列的时延上限为2.0毫秒，由于实际时延大于时延上限，表示队列b中排队的报文数量过多。而导致队列b中排队的报文数量过多的原因，有可能是因为队列b中的部分报文对应的报文流的类别是非时延敏感类。因此，对于这种情况，控制器可以对突发包括的数量对应的预设时间周期进行调整，即对突发包括的报文流的数量的老化周期进行调整，例如增大该老化周期，从而使得报文流的突发包括的报文流的数量能够落入时延非敏感类对应的参数范围内，例如使得报文流包括的突发的数量能够大于表3所示的40，从而使得队列b中的部分报文能够重新被划分到队列c中，从而减缓队列b中排队的报文数量，进一步优化确定报文优先级的结果。此处提及的队列c对应的报文流的类别为时延非敏感类。另外，控制器也可以对非时延敏感类对应的特征参数范围进行调整，例如将非时延敏感类对应的特征参数范围由大于40调整为大于38，从而使得队列b中的部分报文能够重新被划分到队列c中，减缓队列b中排队的报文数量，进一步优化确定报文优先级的结果。
基于以上实施例提供的报文处理方法，本申请实施例还提供了对应的装置，以下结合附图介绍该装置。
参见图4，图4为本申请实施例提供的一种报文处理装置的结构示意图。图4所示的时间报文处理装置400，可以用于执行图2所示的报文处理方法。具体地，该报文处理装置400可以包括获取单元401、第一确定单元402、第二确定单元403和第三确定单元404。
获取单元401，用于获取第一报文；第一确定单元402，用于确定所述第一报文所属的报文流；第二确定单元403，用于当连续获取的属于所述报文流的报文数量大于或者等于第一阈值时，根据所述第一报文确定所述报文流的特征参数，所述连续获取的属于所述报文流的报文中，相邻报文的时间间隔小于或者等于第一时间阈值；第三确定单元404，用于根据所述报文流的特征参数确定所述第一报文的优先级。
在一种实现方式中，所述第三确定单元404具体用于：当所述报文流的特征参数在第一参数范围内时，根据预先确定的特征参数范围与报文优先级的映射关系，将所述第一参数范围对应的报文优先级确定为所述第一报文的优先级，所述特征参数范围与报文优先级的映射关系包括：所述第一参数范围和所述第一参数范围对应的报文优先级的映射关系。
在一种实现方式中，所述第三确定单元404具体用于:获取第一参数的值，所述第一参数的值用于指示第二报文的优先级，所述第二报文与所述第一报文属于同一报文流，所述第二报文为所述第一报文的前一条报文；根据所述第一参数的值和所述报文流的特征参数确定所述第一报文的优先级。
在一种实现方式中，所述装置400还包括第一修改单元，第一修改单元用于对所述第一参数的值进行修改，修改后的第一参数的值用于指示所述第一报文的优先级。
在一种实现方式中，所述装置400还包括第二修改单元，第二修改单元用于修改所述第一报文中第一字段的值，修改后的所述第一字段的值指示所述第一报文的优先级。
在一种实现方式中，所述装置400还包括：转发单元，转发单元用于根据所述第一报文的优先级转发所述第一报文。
在一种实现方式中，所述装置400还包括：第一发送单元，第一发送单元用于将所述报文流的特征参数发送给显示设备进行显示。
在一种实现方式中，所述装置400还包括：第二发送单元和调整单元，第二发送单元用于将所述报文流的特征参数发送给控制器；调整单元用于根据从所述控制器获取的、对多个报文流的特征参数的分类结果，调整对应所述报文流的特征参数范围，调整后的所述特征参数范围用于确定第三报文的优先级，所述第三报文为所述获取单元获取所述第一报文之后获取的报文，所述第三报文属于所述报文流。
在一种实现方式中，所述装置400还包括：第三发送单元，第三发送单元用于将目标队列的实际时延发送给控制器，以使得所述控制器在所述目标队列的实际时延大于所述目标队列的时延上限时对目标参数进行调整，调整后的所述目标参数用于确定第三报文的优先级，所述第三报文为所述网络设备获取所述第一报文之后获取的报文，所述第三报文属于所述报文流；其中：所述目标队列为所述网络设备中用于存储待转发报文的队列；所述目标参数包括以下任意一项或者多项：对应所述报文流的特征参数范围和各个特征参数分别对应的预设时间阈值，第一特征参数为所述各个特征参数中的任意一个特征参数，所述第一特征参数对应的预设时间阈值，用于指示所述网络设备在相邻报文的时间间隔大于所述预设时间阈值时，对所述第一特征参数进行清零。
在一种实现方式中，所述报文流的特征参数，包括以下任意一项或者多项：所述报文流的突发包括的报文数量、所述报文流的最大报文时间间隔，所述报文时间间隔为相邻报文的时间间隔、所述报文流的速率波动系数、以及所述报文流的最大瞬时速率。
由于所述装置400是与以上实施例描述的图2对应的报文处理方法对应的装置，所述装置400的各个单元的具体实现，均与以上方法实施例为同一构思，因此，关于所述装置400的各个单元的具体实现，可以参考以上方法实施例关于图2对应的报文处理方法的描述部分，此处不再赘述。
需要说明的是，前述提及的报文处理装置400，其硬件结构可以为如图5所示的结构，图5为本申请实施例提供的一种报文处理设备的结构示意图。
请参阅图5所示，报文处理设备500包括：处理器510、通信接口520和和存储器530。其中设备500中的处理器510的数量可以一个或多个，图5中以一个处理器为例。本申请实施例中，处理器510、通信接口520和存储器530可通过总线系统或其它方式连接，其中，图5中以通过总线系统540连接为例。
处理器510可以是中央处理器(central processing unit，CPU)，网络处理器(network processor，NP)或者CPU和NP的组合。处理器510还可以进一步包括硬件芯片。上述硬件芯片可以是专用集成电路(application-specific integrated circuit，ASIC)，可编程逻辑器件(programmable logic device，PLD)或其组合。上述PLD可以是复杂可编程逻辑器件 (complex programmable logic device，CPLD)，现场可编程逻辑门阵列(field-programmable gate array，FPGA)，通用阵列逻辑(generic array logic,GAL)或其任意组合。
存储器530可以包括易失性存储器(英文：volatile memory)，例如随机存取存储器(random-access memory，RAM)；存储器530也可以包括非易失性存储器(英文：non-volatile memory)，例如快闪存储器(英文：flash memory)，硬盘(hard disk drive，HDD)或固态硬盘(solid-state drive，SSD)；存储器530还可以包括上述种类的存储器的组合。存储器530例如可以存储前文提及的第一流表以及第二流表。
可选地，存储器530存储有操作系统和程序、可执行模块或者数据结构，或者它们的子集，或者它们的扩展集，其中，程序可包括各种操作指令，用于实现各种操作。操作系统可包括各种系统程序，用于实现各种基础业务以及处理基于硬件的任务。处理器510可以读取存储器530中的程序，实现本申请实施例提供的时间同步方法或者用于时间同步的消息处理方法。
总线系统540可以是外设部件互连标准(peripheral component interconnect，PCI)总线或扩展工业标准结构(extended industry standard architecture，EISA)总线等。总线系统540可以分为地址总线、数据总线、控制总线等。为便于表示，图5中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
本申请实施例还提供了一种计算机可读存储介质，包括指令或计算机程序，当其在计算机上运行时，使得计算机执行以上实施例提供的报文处理方法。
本申请实施例还提供了一种包含指令或计算机程序的计算机程序产品，当其在计算机上运行时，使得计算机执行以上实施例提供的报文处理方法。
本申请中提到的装置或设备，可以是网络设备，比如交换机、路由器，也可以是服务器，也可以是网络设备或服务器的一部分。在一些实施例中，本申请的装置或设备，也可以是部署在设备或网络中的功能模块。
本申请技术方案目前适用于广域网、园区网络和数据中心领域，方法可以用于芯片。
可参见图6a和图6b进行理解，图6a为本申请实施例提供的一种网络架构示意图。图6b为本申请实施例提供的一种数据中心网络的网络架构示意图。
图6a所示的网络架构示意图适用于广域网和园区网络。如图6a所示，客户端607可以通过边缘网络设备606、汇聚层网络设备605和骨干层网络设备604与远端服务器601、数据中心服务器602以及近端服务器603进行交互。本申请实施例提供的技术方案，可以应用于边缘网络设备606、汇聚层网络设备605和骨干层网络设备604上。在图6a所示的网络架构中，数据流量可以包括上行流量和下行流量。上行流量指的是由边缘网络设备606经过汇聚层网络设备605传递给骨干层网络设备604的数据流量。下行流量指的是由骨干层网络设备604经过汇聚层网络设备605传递给边缘网络设备606的数据流量。对于上行流量而言，若边缘网络设备606应用了本申请实施例提供的技术方案，且汇聚层网络设备605和骨干层网络设备604能够正确接受边缘网络设备606所确定的报文优先级，则汇聚层网络设备605和骨干层网络设备604也可以不必应用本申请实施例提供的方案。例如， 边缘网络设备606确定第一报文的优先级之后，对应修改了第一报文的DSCP字段的值，汇聚层网络设备605和骨干层网络设备604可以直接通过该DSCP字段的值确定第一报文的优先级。对于下行流量而言，若骨干层网络设备604应用了本申请实施例提供的技术方案，且汇聚层网络设备605和边缘网络设备606能够正确接受骨干层网络设备604所确定的报文优先级，则汇聚层网络设备605和边缘网络设备606也可以不必应用本申请实施例提供的方案。例如，骨干层网络设备604确定第一报文的优先级之后，对应修改了第一报文的DSCP字段的值，汇聚层网络设备605和边缘网络设备606可以直接通过该DSCP字段的值确定第一报文的优先级。
如图6b所示，数据中心网络可以包括核心层610、汇聚层620和接入层630，核心层610、汇聚层620和接入层630均可以包括多个交换机。其中，汇聚层620中的一个或者多个交换机以及接入层的一个或者多个交换机可以构成一个分组，例如图6b所示的分组640和分组650。数据中心网络的数据交换发生在数据中心网络的各个交换机之间。本申请实施例提供的技术方案，可以应用于图6b所示的各个交换机。类似地，若接入层630的交换机应用了本申请实施例提供的技术方案，且汇聚层620的交换机和核心层610的交换机能够正确接受接入层630的交换机所确定的报文优先级，则汇聚层620的交换机和核心层610的交换机也可以不必应用本申请实施例提供的方案。
本申请的技术方案，作为一种流识别分类技术，可以作用在设备入口侧的流量管理模块之前。例如，可参见图7a和图7b进行理解，图7a和图7b为本申请实施例提供一种示例性应用场景示意图。如图7a所示，本申请提供的流识别分类技术，可以嵌入在流量管理模块701之前的其它模块702上。如图7b所示，本申请提供的流识别分类技术应用在一个独立的流分类模块703上，该独立的流分类模块703位于流量管理模块701和其它模块702之间。本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”、“第四”等(如果存在)是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统，装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，单元的划分，仅仅为一种逻辑业务划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
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单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各业务单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件业务单元的形式实现。
集成的单元如果以软件业务单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
本领域技术人员应该可以意识到，在上述一个或多个示例中，本发明所描述的业务可以用硬件、软件、固件或它们的任意组合来实现。当使用软件实现时，可以将这些业务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。
以上的具体实施方式，对本发明的目的、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，以上仅为本发明的具体实施方式而已。
以上，以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (23)

  1. 一种报文处理方法，其特征在于，包括：
    获取第一报文，确定所述第一报文所属的报文流；
    当连续获取的属于所述报文流的报文数量大于或者等于第一阈值时，根据所述第一报文确定所述报文流的特征参数，所述连续获取的属于所述报文流的报文中，相邻报文的时间间隔小于或者等于第一时间阈值；
    根据所述报文流的特征参数确定所述第一报文的优先级。
  2. 根据权利要求1所述的方法，其特征在于，根据所述报文流的特征参数确定所述第一报文的优先级，包括：
    当所述报文流的特征参数在第一参数范围内时，根据预先确定的特征参数范围与报文优先级的映射关系，将所述第一参数范围对应的报文优先级确定为所述第一报文的优先级，所述特征参数范围与报文优先级的映射关系包括：所述第一参数范围和所述第一参数范围对应的报文优先级的映射关系。
  3. 根据权利要求1所述的方法，其特征在于，根据所述报文流的特征参数确定所述第一报文的优先级，包括:
    获取第一参数的值，所述第一参数的值用于指示第二报文的优先级，所述第二报文与所述第一报文属于同一报文流，所述第二报文为所述第一报文的前一条报文；
    根据所述第一参数的值和所述报文流的特征参数确定所述第一报文的优先级。
  4. 根据权利要求3所述的方法，其特征在于，还包括：
    对所述第一参数的值进行修改，修改后的第一参数的值用于指示所述第一报文的优先级。
  5. 根据权利要求1-4任意一项所述的方法，其特征在于，还包括：
    修改所述第一报文中第一字段的值，修改后的所述第一字段的值指示所述第一报文的优先级。
  6. 根据权利要求1-5任意一项所述的方法，其特征在于，还包括：
    根据所述第一报文的优先级转发所述第一报文。
  7. 根据权利要求1-6所述的方法，其特征在于，还包括：
    将所述报文流的特征参数发送给显示设备进行显示。
  8. 根据权利要求1-7任意一项所述的方法，其特征在于，还包括：
    将所述报文流的特征参数发送给控制器，并根据从所述控制器获取的、对多个报文流的特征参数的分类结果，调整对应所述报文流的特征参数范围，调整后的所述对应所述报文流的特征参数范围用于确定第三报文的优先级，所述第三报文为获取所述第一报文之后获取的报文，所述第三报文属于所述报文流。
  9. 根据权利要求1-7任意一项所述的方法，其特征在于，还包括：
    将目标队列的实际时延发送给控制器，以使得所述控制器在所述目标队列的实际时延大于所述目标队列的时延上限时对目标参数进行调整，调整后的所述目标参数用于确定第三报文的优先级，所述第三报文为所述网络设备获取所述第一报文之后获取的报文，所述第三报文属于所述报文流；
    其中：
    所述目标队列为包含待转发报文的队列；所述目标参数包括以下任意一项或者多项：对应所述报文流的特征参数范围和各个特征参数分别对应的预设时间阈值，第一特征参数为所述各个特征参数中的任意一个特征参数，所述第一特征参数对应的预设时间阈值，用于指示所述网络设备在相邻报文的时间间隔大于所述预设时间阈值时，对所述第一特征参数进行清零。
  10. 根据权利要求1-9任意一项所述的方法，其特征在于，所述报文流的特征参数，包括以下任意一项或者多项：
    所述报文流的突发包括的报文数量、所述报文流的最大报文时间间隔，所述报文时间间隔为相邻报文的时间间隔、所述报文流的速率波动系数、以及所述报文流的最大瞬时速率。
  11. 一种报文处理装置，其特征在于，包括：
    获取单元，用于获取第一报文；
    第一确定单元，用于确定所述第一报文所属的报文流；
    第二确定单元，用于当连续获取的属于所述报文流报文数量大于或者等于第一阈值时，根据所述第一报文确定所述报文流的特征参数，所述连续获取的属于所述报文流的报文中，相邻报文的时间间隔小于或者等于第一时间阈值；
    第三确定单元，用于根据所述报文流的特征参数确定所述第一报文的优先级。
  12. 根据权利要求11所述的装置，其特征在于，所述第三确定单元，具体用于：
    当所述报文流的特征参数在第一参数范围内时，根据预先确定的特征参数范围与报文优先级的映射关系，将所述第一参数范围对应的报文优先级确定为所述第一报文的优先级，所述特征参数范围与报文优先级的映射关系包括：所述第一参数范围和所述第一参数范围对应的报文优先级的映射关系。
  13. 根据权利要求11所述的装置，其特征在于，所述第三确定单元，具体用于:
    获取第一参数的值，所述第一参数的值用于指示第二报文的优先级，所述第二报文与所述第一报文属于同一报文流，所述第二报文为所述第一报文的前一条报文；
    根据所述第一参数的值和所述报文流的特征参数确定所述第一报文的优先级。
  14. 根据权利要求13所述的装置，其特征在于，还包括：
    第一修改单元，用于对所述第一参数的值进行修改，修改后的第一参数的值用于指示所述第一报文的优先级。
  15. 根据权利要求11-14任意一项所述的装置，其特征在于，还包括：
    第二修改单元，用于修改所述第一报文中第一字段的值，修改后的所述第一字段的值指示所述第一报文的优先级。
  16. 根据权利要求11-15任意一项所述的装置，其特征在于，还包括：
    转发单元，用于根据所述第一报文的优先级转发所述第一报文。
  17. 根据权利要求11-16所述的装置，其特征在于，还包括：
    第一发送单元，用于将所述报文流的特征参数发送给显示设备进行显示。
  18. 根据权利要求11-17任意一项所述的装置，其特征在于，还包括：
    第二发送单元，用于将所述报文流的特征参数发送给控制器；
    调整单元，用于根据从所述控制器获取的、对多个报文流的特征参数的分类结果，调整对应所述报文流的特征参数范围，调整后的所述对应所述报文流的特征参数范围用于确定第三报文的优先级，所述第三报文为所述获取单元获取所述第一报文之后获取的报文，所述第三报文属于所述报文流。
  19. 根据权利要求11-17任意一项所述的装置，其特征在于，所述装置还包括：
    第三发送单元，用于将目标队列的实际时延发送给控制器，以使得所述控制器在所述目标队列的实际时延大于所述目标队列的时延上限时对目标参数进行调整，调整后的所述目标参数用于确定第三报文的优先级，所述第三报文为所述网络设备获取所述第一报文之后获取的报文，所述第三报文属于所述报文流；
    其中：
    所述目标队列为所述网络设备中用于存储待转发报文的队列；所述目标参数包括以下任意一项或者多项：对应所述报文流的特征参数范围和各个特征参数分别对应的预设时间阈值，第一特征参数为所述各个特征参数中的任意一个特征参数，所述第一特征参数对应的预设时间阈值，用于指示所述网络设备在相邻报文的时间间隔大于所述预设时间阈值时，对所述第一特征参数进行清零。
  20. 根据权利要求11-19任意一项所述的装置，其特征在于，所述报文流的特征参数，包括以下任意一项或者多项：
    所述报文流的突发包括的报文数量、所述报文流的最大报文时间间隔，所述报文时间间隔为相邻报文的时间间隔、所述报文流的速率波动系数、以及所述报文流的最大瞬时速率。
  21. 一种设备，其特征在于，包括：处理器和存储器；
    所述存储器，用于存储指令或计算机程序；
    所述处理器，用于执行所述指令或计算机程序，执行权利要求1-10任意一项所述的方法。
  22. 一种计算机可读存储介质，其特征在于，包括指令或计算机程序，当其在计算机上运行时，使得计算机执行以上权利要求1-10任意一项所述的方法。
  23. 一种计算机程序产品，其特征在于，包括指令或计算机程序，当其在计算机上运行时，使得计算机执行以上权利要求1-10任意一项所述的方法。
PCT/CN2020/119154 2019-12-20 2020-09-30 一种报文处理方法及装置 WO2021120783A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911329488.1 2019-12-20
CN201911329488.1A CN113014508A (zh) 2019-12-20 2019-12-20 一种报文处理方法及装置

Publications (1)

Publication Number Publication Date
WO2021120783A1 true WO2021120783A1 (zh) 2021-06-24

Family

ID=76381862

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/119154 WO2021120783A1 (zh) 2019-12-20 2020-09-30 一种报文处理方法及装置

Country Status (2)

Country Link
CN (1) CN113014508A (zh)
WO (1) WO2021120783A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113872733A (zh) * 2021-09-29 2021-12-31 天翼物联科技有限公司 报文重发方法、装置、计算机设备及计算机可读存储介质
CN114006867A (zh) * 2021-09-29 2022-02-01 苏州浪潮智能科技有限公司 报文传输优先级控制方法、系统、装置及存储介质
CN114945005A (zh) * 2022-05-24 2022-08-26 湖南快乐阳光互动娱乐传媒有限公司 一种消息处理方法及相关设备
CN115571072A (zh) * 2022-10-24 2023-01-06 中国第一汽车股份有限公司 抬头显示器模式转换方法、装置、电子设备及存储介质
WO2023168688A1 (zh) * 2022-03-11 2023-09-14 华为技术有限公司 发送数据的方法和相关设备

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115529638A (zh) * 2021-06-25 2022-12-27 华为技术有限公司 一种报文处理方法及相关设备
CN114598752B (zh) * 2022-03-07 2023-12-15 潍柴动力股份有限公司 报文处理方法、装置以及相关设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101815357A (zh) * 2010-04-23 2010-08-25 华为技术有限公司 无线承载内数据发送的方法及装置
US20140226477A1 (en) * 2011-10-19 2014-08-14 Huawei Technologies Co., Ltd. Method, Apparatus, and System for Improving user Experience of Small Flow User
CN104579962A (zh) * 2015-01-23 2015-04-29 盛科网络（苏州）有限公司 一种区分不同报文的QoS策略的方法及装置
CN107342906A (zh) * 2016-04-29 2017-11-10 华为技术有限公司 一种大象流的检测方法、设备及系统
CN108322405A (zh) * 2018-03-21 2018-07-24 山东超越数控电子股份有限公司 一种基于数据的流量均衡方法与装置
CN110290178A (zh) * 2019-05-30 2019-09-27 厦门网宿有限公司 一种数据流的调度方法、电子设备及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101815357A (zh) * 2010-04-23 2010-08-25 华为技术有限公司 无线承载内数据发送的方法及装置
US20140226477A1 (en) * 2011-10-19 2014-08-14 Huawei Technologies Co., Ltd. Method, Apparatus, and System for Improving user Experience of Small Flow User
CN104579962A (zh) * 2015-01-23 2015-04-29 盛科网络（苏州）有限公司 一种区分不同报文的QoS策略的方法及装置
CN107342906A (zh) * 2016-04-29 2017-11-10 华为技术有限公司 一种大象流的检测方法、设备及系统
CN108322405A (zh) * 2018-03-21 2018-07-24 山东超越数控电子股份有限公司 一种基于数据的流量均衡方法与装置
CN110290178A (zh) * 2019-05-30 2019-09-27 厦门网宿有限公司 一种数据流的调度方法、电子设备及存储介质

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113872733A (zh) * 2021-09-29 2021-12-31 天翼物联科技有限公司 报文重发方法、装置、计算机设备及计算机可读存储介质
CN114006867A (zh) * 2021-09-29 2022-02-01 苏州浪潮智能科技有限公司 报文传输优先级控制方法、系统、装置及存储介质
CN114006867B (zh) * 2021-09-29 2023-08-29 苏州浪潮智能科技有限公司 报文传输优先级控制方法、系统、装置及存储介质
CN113872733B (zh) * 2021-09-29 2024-01-05 天翼物联科技有限公司 报文重发方法、装置、计算机设备及计算机可读存储介质
WO2023168688A1 (zh) * 2022-03-11 2023-09-14 华为技术有限公司 发送数据的方法和相关设备
CN114945005A (zh) * 2022-05-24 2022-08-26 湖南快乐阳光互动娱乐传媒有限公司 一种消息处理方法及相关设备
CN114945005B (zh) * 2022-05-24 2024-02-06 湖南快乐阳光互动娱乐传媒有限公司 一种消息处理方法及相关设备
CN115571072A (zh) * 2022-10-24 2023-01-06 中国第一汽车股份有限公司 抬头显示器模式转换方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
CN113014508A (zh) 2021-06-22

Similar Documents

Publication Publication Date Title
WO2021120783A1 (zh) 一种报文处理方法及装置
US10659342B2 (en) Flow entry configuration method, apparatus, and system
EP3661137B1 (en) Packet control method and network device
US11032205B2 (en) Flow control method and switching device
US20200358701A1 (en) Packet Sending Method and Apparatus, and Storage Device
CN111316605A (zh) 第3层公平速率拥塞控制通知
EP3852323A1 (en) Congestion control method, and network apparatus
US11870698B2 (en) Congestion control method and apparatus, communications network, and computer storage medium
JP2021516012A (ja) ネットワークにおけるフロー管理
US9571410B2 (en) Credit-based link level flow control and credit exchange using DCBX
CN106330742B (zh) 一种流量控制的方法及网络控制器
WO2018108093A1 (zh) 拥塞控制方法及装置和计算机存储介质
WO2015172668A1 (zh) 网络中拥塞窗口的确定方法和装置
US11653251B2 (en) Packet forwarding method and apparatus
US11165705B2 (en) Data transmission method, device, and computer storage medium
WO2022028342A1 (zh) 一种拥塞流的处理方法及设备
WO2021179648A1 (zh) 一种光纤通信中的数据转发方法以及相关装置
EP2854346A1 (en) Method for adjusting bandwidth in network virtualization system
WO2021012902A1 (zh) 一种处理网络拥塞的方法以及相关装置
WO2017016300A1 (zh) 令牌申请处理方法、装置和计算机存储介质
CN110336759B (zh) 基于rdma的协议报文转发方法及装置
US20170373982A1 (en) System and method for mtu size reduction in a packet network
CN113765796A (zh) 流量转发控制方法及装置
US20190140965A1 (en) Method for obtaining path information of data packet and device
WO2023246616A1 (zh) 网络拥塞控制方法及装置

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: 20901191

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20901191

Country of ref document: EP

Kind code of ref document: A1