WO2021208682A1 - 网络设备的数据采样方法、装置、设备及介质 - Google Patents

网络设备的数据采样方法、装置、设备及介质 Download PDF

Info

Publication number
WO2021208682A1
WO2021208682A1 PCT/CN2021/082475 CN2021082475W WO2021208682A1 WO 2021208682 A1 WO2021208682 A1 WO 2021208682A1 CN 2021082475 W CN2021082475 W CN 2021082475W WO 2021208682 A1 WO2021208682 A1 WO 2021208682A1
Authority
WO
WIPO (PCT)
Prior art keywords
sampling
report
value
message
interval
Prior art date
Application number
PCT/CN2021/082475
Other languages
English (en)
French (fr)
Inventor
杨明杰
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to EP21788665.4A priority Critical patent/EP4138340A4/en
Priority to US17/914,670 priority patent/US20230131524A1/en
Publication of WO2021208682A1 publication Critical patent/WO2021208682A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/06—Generation of reports
    • H04L43/067—Generation of reports using time frame reporting
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/02—Capturing of monitoring data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
    • G06F11/3006—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3065—Monitoring arrangements determined by the means or processing involved in reporting the monitored data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3089—Monitoring arrangements determined by the means or processing involved in sensing the monitored data, e.g. interfaces, connectors, sensors, probes, agents
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
    • G06F11/3452—Performance evaluation by statistical analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/02—Capturing of monitoring data
    • H04L43/022—Capturing of monitoring data by sampling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/02—Capturing of monitoring data
    • H04L43/022—Capturing of monitoring data by sampling
    • H04L43/024—Capturing of monitoring data by sampling by adaptive sampling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/06—Generation of reports

Definitions

  • This application relates to the field of communications technology, for example, to a data sampling method, device, device, and medium for network equipment.
  • the traditional message statistics sampling method is: when the message passes through the network device, the network processor (Network Processing Unit, NPU) of the network device performs the message statistics, and the NPU stores multiple statistical values in the NPU.
  • the central processing unit (CPU) of the device regularly sends query messages to the NPU, samples the obtained statistical values (as shown in Fig. 1) at regular intervals, and presents them to the man-machine interface after processing.
  • the CPU needs to process multiple query messages, which seriously consumes the processing resources of the CPU and the bandwidth of the configuration query channel between the CPU and the NPU.
  • the CPU cannot process so many messages in a short time. Even if a high-performance CPU can process so many messages, it will consume a lot of processing resources of the high-performance CPU, and the price of the high-performance CPU Very expensive.
  • This application provides a data sampling method, device, device, and medium for network equipment, so as to greatly reduce the occupied CPU processing resources on the premise of meeting the requirement of high sampling rate of network equipment data.
  • a data sampling method for network equipment is provided, which is applied to network equipment, including:
  • the sampling parameter of the target data is configured to the coprocessor through the CPU; the sampling instruction message is generated by the coprocessor at a set frequency according to the sampling parameter, and the sampling instruction message is sent to the NPU; The sampling instruction message generates a sampling report message of the target data, and sends the sampling report message to the CPU.
  • a data sampling method for network devices is also provided, which is applied to the CPU, and includes:
  • the sampling parameter of the target data is configured to the coprocessor, where the sampling parameter is used by the coprocessor to generate a sampling instruction message at a set frequency and send the sampling instruction message to the NPU, and the sampling instruction The message is used by the NPU to generate a sampling report message of the target data and send the sampling report message to the CPU; and receive a sampling report message of the target data sent by the NPU.
  • a data sampling method for network equipment is also provided, which is applied to a coprocessor, and includes:
  • the NPU generates a sampling report message of the target data and sends the sampling report message to the CPU.
  • a data sampling method for network equipment is also provided, which is applied to the NPU, including:
  • sampling instruction message sent by the coprocessor, where the sampling instruction message is generated and sent by the coprocessor at a set frequency according to the sampling parameters of the target data sent by the CPU; according to the sampling instruction message A sampling report message of the target data is generated, and the sampling report message is sent to the CPU.
  • a data sampling device for network equipment is also provided, which is applied to the CPU, and includes:
  • the sampling parameter configuration module is configured to configure the sampling parameters of the target data to the coprocessor, wherein the sampling parameters are used by the coprocessor to generate a sampling instruction message at a set frequency and send the sampling instruction message To the NPU, the sampling instruction message is used by the NPU to generate a sampling report message of the target data and send the sampling report message to the CPU; the sampling report message receiving module is configured to receive the NPU transmission Sampling report message of the target data.
  • a data sampling device for network equipment is also provided, which is applied to a coprocessor, and includes:
  • the sampling parameter reading module is set to read the sampling parameters of the target data configured by the CPU;
  • the sampling instruction message sending module is set to generate a sampling instruction message at a set frequency according to the sampling parameter, and send the sampling instruction message
  • the message is sent to the NPU, where the sampling instruction message is used by the NPU to generate a sampling report message of the target data and send the sampling report message to the CPU.
  • a data sampling device for network equipment is also provided, which is applied to the NPU, and includes:
  • the sampling instruction message receiving module is configured to receive the sampling instruction message sent by the coprocessor, where the sampling instruction message is generated and sent by the coprocessor at a set frequency according to the sampling parameters of the target data sent by the CPU
  • the sampling report message sending module is configured to generate a sampling report message of the target data according to the sampling instruction message, and send the sampling report message to the CPU.
  • a CPU including: one or more processing units; a storage unit for storing one or more programs; when the one or more programs are executed by the one or more processing units, the one Or multiple processing units implement the above-mentioned data sampling method applied to the network device in the CPU.
  • a coprocessor which includes: one or more processing units; a storage unit for storing one or more programs; when the one or more programs are executed by the one or more processing units, the The one or more processing units implement the above-mentioned data sampling method applied to the network device in the coprocessor.
  • An NPU including: one or more processing units; a storage unit for storing one or more programs; when the one or more programs are executed by the one or more processing units, the one Or multiple processing units implement the above-mentioned data sampling method applied to the network device in the NPU.
  • a network device including: a CPU, a coprocessor, and an NPU, wherein the CPU executes the above-mentioned data sampling method applied to the network device in the CPU, and the coprocessor executes the above-mentioned application to the coprocessor In the data sampling method of the network device in the NPU, the NPU executes the above-mentioned data sampling method applied to the network device in the NPU.
  • a storage medium is also provided, the storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned network device data sampling method is realized.
  • Figure 1 is a schematic diagram of the module structure of a traditional network device
  • FIG. 2 is a schematic diagram of a module structure of a network device provided by an embodiment of the application.
  • FIG. 3 is a schematic flowchart of a data sampling method for a network device according to an embodiment of this application;
  • FIG. 4 is a schematic flowchart of another data sampling method for network equipment provided by an embodiment of this application.
  • FIG. 5 is a schematic flowchart of another data sampling method for network equipment according to an embodiment of the application.
  • FIG. 6 is a schematic flowchart of another data sampling method for network equipment according to an embodiment of the application.
  • FIG. 7 is a schematic flowchart of another data sampling method for network equipment according to an embodiment of the application.
  • FIG. 8 is a schematic diagram of the module structure of a data sampling device for network equipment provided by an embodiment of the application.
  • FIG. 9 is a schematic diagram of the module structure of another data sampling device for network equipment according to an embodiment of the application.
  • FIG. 10 is a schematic diagram of the module structure of another data sampling apparatus for network equipment according to an embodiment of the application.
  • FIG. 11 is a schematic structural diagram of a CPU provided by an embodiment of the application.
  • FIG. 12 is a schematic structural diagram of a coprocessor provided by an embodiment of the application.
  • FIG. 13 is a schematic structural diagram of an NPU provided by an embodiment of the application.
  • the network equipment provided by this application includes a CPU, a coprocessor, and an NPU.
  • the CPU is set to process user configuration and sample result processing and display, and assume the overall control role in the system;
  • NPU is the core device of network equipment, has certain programming capabilities, and can perform high-speed processing of network messages. In this application , NPU will perform various statistics on the packets after receiving the packets on the network.
  • the coprocessor can be an external Field Programmable Gate Array (FPGA) or a packet generator inside the NPU, which has simple calculation processing capabilities and fast packet sending capabilities.
  • FPGA Field Programmable Gate Array
  • the CPU configures the sampling parameters of the target data to the coprocessor.
  • the coprocessor After receiving the sampling parameters, the coprocessor generates a sampling instruction message at a set frequency according to the sampling parameters and sends it to the NPU, and the NPU sends it to the NPU according to the sampling instruction message.
  • a sampling report message of the target data is generated and sent to the CPU so that the CPU can process the sampling result carried in the sampling report message after receiving the sampling report message.
  • FIG. 3 is a schematic flowchart of a data sampling method for a network device according to an embodiment of the application.
  • This method is suitable for sampling the business (or performance) statistics, real-time alarms, real-time status, etc. processed by the network equipment at a high frequency.
  • the method can be executed by the data sampling device applied to the network device in the network device provided in this application, and the data sampling device of the network device can be implemented by software and/or hardware and integrated in the network device.
  • the data sampling method applied to the network equipment in the network equipment provided by this application includes:
  • Target data refers to data that needs to be sampled on network equipment, such as business statistics processed by the equipment, real-time alarm data, and real-time status data.
  • the data sampling method described in this embodiment involves statistics such as network device port message sending and receiving, multi-level services, access control list (Access Control List, ACL), and sampling of multiple alarms and multiple states.
  • statistics such as network device port message sending and receiving, multi-level services, access control list (Access Control List, ACL), and sampling of multiple alarms and multiple states.
  • ACL Access Control List
  • the user configuration parameters are preset to the CPU.
  • the sampling parameters are obtained and configured to the coprocessor.
  • the coprocessor is an external FPGA, or a packet generator in the NPU, which is not limited in this embodiment, and devices with simple calculation processing capabilities and fast packet sending capabilities in network equipment can all be used as co-processing Device.
  • User configuration parameters can be sampling function items, sampling frequency, report frequency, etc.
  • the report frequency may be the sample value report frequency and/or the average value report frequency.
  • User configuration parameters can also include whether to suppress redundancy.
  • the average value may refer to the average value of all sampled values in a period.
  • the average value may refer to the average value of the maximum sample value in the period and the minimum sample value in the period, that is, the maximum and minimum average values.
  • sampling parameters at least include: sampling number, sampling function item, sampling frequency, and reporting interval.
  • Sampling number used to uniquely identify the sampling task
  • sampling function item used to identify the name of the sampling task, such as business statistics 1, business statistics 2, alarm 1, alarm 2, state 1, state 2, etc.
  • sampling frequency is set by the user
  • the parameter defines the number of data taken out of continuous data per second, also known as the sampling rate
  • the reporting interval defines the number of samples to be reported once.
  • the sampling value report interval is determined according to the report frequency and the sampling frequency.
  • the reporting interval includes: sampling value reporting interval and/or average value reporting interval.
  • sampling value report interval defines the data sampling value to be reported every sampling times
  • average reporting interval defines the data average value to be reported every sampling times.
  • the sampling value report interval is determined according to the sampling value report frequency and the sampling frequency.
  • the average value report interval is determined according to the average value report frequency and the sampling frequency.
  • the CPU calculates the reporting interval according to the sampling frequency and the reporting frequency.
  • sampling parameters include at least: sampling number, sampling function item, sampling frequency, reporting interval, and suppressing redundancy identification.
  • Redundancy suppression flag a parameter set by the user, used to identify whether the redundancy suppression function is required during the sampling process.
  • the suppression of redundancy is identified as a valid value, which corresponds to the need for redundancy suppression in the sampling process.
  • sampling parameters configured by the CPU to the coprocessor may be at least one group. That is, the CPU can configure corresponding sampling parameters to the coprocessor for multiple sampling function items.
  • S120 Generate a sampling instruction message at a set frequency according to the sampling parameter by the coprocessor, and send the sampling instruction message to the NPU.
  • the coprocessor After the coprocessor reads the sampling parameter configured by the CPU, it generates a sampling instruction message at a set frequency according to the sampling parameter and sends it to the NPU.
  • the set frequency can be any preset frequency, for example, it can be the operating frequency of the coprocessor.
  • the sampling instruction message generated by the coprocessor at a set frequency according to the sampling parameter may be: the sampling function that the sampling frequency is not zero is performed by the coprocessor according to the sampling parameter Item, the sampling instruction message is generated at the sampling frequency.
  • the set frequency may be the sampling frequency in the sampling parameter. That is, the coprocessor generates a sampling instruction message at the sampling frequency and sends it to the NPU, so that the NPU obtains the corresponding sampling value according to the sampling instruction message.
  • a logic table may be maintained inside the coprocessor, which may be referred to as a sampling parameter configuration table here, which is used to describe the sampling function items that need to be performed currently and the corresponding parameters.
  • the quick sampling parameter configuration table is a logical table that the coprocessor can access. It can be configured by the CPU, that is, the CPU can configure the sampling parameters into the sampling parameter configuration table, so as to control the coprocessor based on the sampling parameter configuration table. Sampling parameters are used to send sampling instructions.
  • the quick sampling parameter configuration table may be as shown in Table 1.
  • the sampling instruction message carries a message type, sampling number, sampling function item, report count value, and report interval.
  • the message type carried in the sampling instruction message is the sampling type.
  • the report count value refers to the number of sampling indication messages that have been sent for the current report, and its variation range is from the initial value to the reporting interval.
  • the report count value starts from the initial value (for example, from 1). If the report count value is not equal to the report interval, the report count value in the next sample indication message sent by the coprocessor is incremented by 1, if the report count value is equal to Report interval, the report count value in the next sampling instruction message sent by the coprocessor is reset to the initial value (that is, 1).
  • the report count value is the sampling value report count value.
  • the sampling value report count value is sequentially increased from the initial value. Reset to the initial value when the sampling value reporting interval is reached.
  • sampling value report count value starts from the initial value (for example, starting from 1). If the sampling value report count value is not equal to the sampling value report interval, the next sampling indication message sent by the coprocessor will increase the sampling value report count value 1. If the sampling value report count value is equal to the sampling value report interval, the sampling value report count value in the next sampling indication message sent by the coprocessor is reset to the initial value (that is, 1).
  • the report count value is the average value report count value; for the sampling function items whose average value report interval is not zero, in the sampling instruction messages sent sequentially , The average value report count value is sequentially increased from the initial value, and reset to the initial value when it increases to the average value report interval.
  • the average value report count value in the sampling indication message starts from the initial value (for example, from 1 Start), if the average value report count value is not equal to the average value report interval, the next sample indication message sent by the coprocessor will increase the average value report count value by 1. If the average value report count value is equal to the average value report interval, Then the average report count value in the next sampling instruction message sent by the coprocessor is reset to the initial value (that is, 1).
  • the significance of setting the report count value is that the NPU does not have the timing counting function, and it does not know when to send a sampling report message to the CPU.
  • the NPU can compare the sampling indication message with Report the count value and the report interval to determine whether a sampling report message needs to be sent to the CPU.
  • the sampling instruction message carries message type, sampling number, sampling function item, report count value, report interval, and redundancy suppression identifier.
  • the redundancy suppression flag is used to instruct the NPU to implement the redundancy suppression function when the suppression redundancy flag is a valid value.
  • the NPU implements the redundancy suppression function, if the current sampling data changes, it will immediately report to the CPU; if the current sampling data does not change, it will report to the CPU at the reporting interval.
  • the sampling instruction message further includes: a total number of sampling function items, where the total number of sampling function items is used to indicate the sampling number and the number of groups of sampling function items included in the sampling instruction message; The sampling frequency and reporting interval of the sampling function items in different groups are the same.
  • sampling function items are related and have the same sampling frequency and report frequency, they can be configured at the same time. Furthermore, the coprocessor can carry multiple sampling function items in a sampling instruction message, and use the sampling instruction The message carries the total number of sampling function items.
  • sampling number and the number of groups of sampling function items included in the sampling indication message are at least one group. When only one sampling function item is carried in the sampling instruction message, the total number of sampling function items is 1.
  • Table 2 shows the field information included in a sampling indication message.
  • S130 Generate a sampling report message of the target data by the NPU according to the sampling instruction message, and send the sampling report message to the CPU.
  • the NPU is the core device of the network device and has a certain programming capability, capable of high-speed processing of network messages. In this application, after receiving the message sent by the coprocessor, the NPU will perform statistical analysis on various messages.
  • the NPU can access to read and write a piece of memory space, which can be called a sampling memory space, and the sampling memory space is set to store sample values.
  • Table 3 shows a sample value storage table in the sample memory space, where the average value may be the maximum and minimum average values.
  • the average value is the maximum and minimum average value as an example for explanation, that is, the average value mentioned below can be the maximum and minimum average value.
  • sampling report message of the target data generated by the NPU according to the sampling indication message may be:
  • the NPU updates the locally stored sampling value according to the sampling instruction message; when the report count value in the sampling instruction message is equal to the reporting interval, the target data is generated according to the locally stored sampling value Sampling report message; wherein, the sampling report message includes a sampling value report message and/or an average value report message.
  • the NPU moves the sampling value in the corresponding register to the sampling memory space in sequence, which can be added to the sampling value storage table shown in Table 3, and the sampling value storage table The sampling value of is used to fill the sampling report message.
  • the NPU moves the sampling value in the corresponding register to the sampling memory space in sequence according to the fields in the received sampling instruction message, and can calculate the maximum value in the period that matches the sampling number. And the minimum value in the period is updated. Furthermore, when the average value needs to be reported, the calculation can be performed directly based on the maximum value within the period and the minimum value within the period that match the sampling number in the sampling value storage table shown in Table 3.
  • the average value is the average value of the maximum value in the period and the minimum value in the period.
  • the NPU compares the report count value in the sampling indication message with the reporting interval to determine whether the sampling report message transmission period has come.
  • sampling value report count value in the sampling indication message is not equal to the sampling value report interval, it means that the sampling report message sending cycle about the sampling value has not been reached; if the sampling value report count value and the sampling value in the sampling indication message If the reporting interval is equal, it means that the sampling report message sending cycle about the sampling value has come, and the NPU generates a corresponding sampling report message and sends it to the CPU.
  • the average value report interval in the sampling indication message is zero, it means that the average value report is not required; if the average value report interval in the sampling indication message is not zero, it means that the average value report is required. If the sampling indication message is reported The average value report count value in the text is not equal to the average value report interval, it means that the sampling report message sending cycle about the average value has not been reached; if the average value report count value in the sampling indication message is equal to the average value report interval, It indicates that the sampling report message sending cycle about the average value has arrived, and the NPU generates a corresponding sampling report message and sends it to the CPU.
  • sampling report message of the target data generated by the NPU according to the sampling indication message may be:
  • the NPU updates the locally stored sampling value according to the sampling instruction message, and determines whether the sampling value at the current moment has changed; if it is determined by the NPU If the sampling value at the current time changes, a sampling report message of the target data is generated according to the locally stored sampling value; if it is determined through the NPU that the sampling value at the current time has not changed, when the sampling instruction message is When the report count value is equal to the report interval, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; wherein, the sampling report message includes a sampling value report message and / Or average report message.
  • the NPU updates the locally stored sampling value according to the sampling indication message, and determines whether the sampling value at the current time has changed: if the current time If the sampling value changes, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; if the sampling value at the current moment does not change, when all the sampling indication messages are When the report count value is equal to the report interval, a sampling report message of the target data is generated according to the locally stored sampling value, and sent to the CPU.
  • sampling value at the current time When judging whether the sampling value at the current time has changed, for statistical values, if the sampling value at the current time is not zero, it is considered that the sampling value at the current time has changed. For the alarm value and status value, if the sampling value at the current time is If it is inconsistent with the sampled value at the previous moment, it is considered that the sampled value at the current moment has changed. Furthermore, when the suppression redundancy flag is a valid value, if the sampling value at the current moment has changed, the NPU will immediately send a sampling report message to the CPU. If the sampling value at the current moment has not changed, the NPU will not report to the CPU until the reporting period. Send sampling report message.
  • Table 4 shows the field information contained in a sampling report message about sampling values.
  • Table 5 shows the field information contained in a sample report message about the average value.
  • the CPU analyzes the received sampling report message sent by the NPU, obtains the sampling value carried in the sampling report message, and performs setting processing on the sampling value to present it to the man-machine interface.
  • the technical solution provided by this embodiment can achieve a higher sampling frequency (the statistical data can be sampled at a frequency faster than 100 times per second at the fastest) and will not consume a lot of money.
  • the CPU processing resources also save the bandwidth of the message channel between the CPU and the NPU. Furthermore, the statistical monitoring performance of the entire network data is improved, and the reliability of the telecommunication service is guaranteed.
  • the maximum and minimum average value report is not required, and the redundancy suppression function is not required.
  • the user parameters are preset to the CPU, the average reporting interval is set to 0, and the suppression redundancy flag is set to a non-true value.
  • the CPU calculates the sampling value report interval.
  • the CPU calculates and obtains an unoccupied sampling number, and configures the sampling parameters to the coprocessor.
  • the sampling parameters include the sampling number and sampling Frequency, sampling function items, sampling value report interval, average value report interval, redundant identification suppression, etc.
  • the coprocessor According to the sampling parameters configured by the CPU, the coprocessor generates a sampling number, sampling function item, sampling value report count value, sampling value report interval, and average value report for the sampling function items whose sampling frequency is not 0 at the configured sampling frequency.
  • the count value, the average reporting interval, and the sampling indication message of the suppression redundancy flag are sent to the NPU.
  • the sampling value report count value can start from 1, each time a sampling indication message is sent, if the sampling value report count value corresponding to the sampling function item is not equal to the corresponding sampling value report interval, the sampling in the next sampling indication message sent The sampling value report count value corresponding to the function item is incremented by 1; if the sampling value report count value corresponding to the sampling function item is equal to the corresponding sampling value report interval, the sampling value report corresponding to the sampling function item in the next sampling indication message sent The count value is set to 1.
  • the average value report interval corresponding to the sampling function item is 0, and the average value report count value in the sampling indication message can be a null value or other values, which is not limited.
  • sampling function items are related, they have the same sampling frequency and report frequency and can be configured at the same time. Multiple sampling function items can be carried in the sampling instruction message at the same time, and the sampling function can be carried in the sampling instruction message The total number of items field.
  • the NPU After receiving the sampling instruction message, the NPU recognizes the sampling function items in the sampling instruction message, and reads the corresponding statistics, alarms, and status values.
  • the NPU judges that the suppression redundancy flag is not a true value, it moves the sample value to the matched sample value sequence according to the sampling instruction message, and updates the maximum value and the minimum value in the period; the NPU according to the sampling instruction message If the average value report interval is zero, it is judged that the maximum and minimum average values do not need to be reported; the NPU judges whether the sampling value report count value and the sampling value report interval in the sampling indication message are equal, if the sampling value report count value and the sampling value report interval are not equal , Indicating that the sampling value report message sending cycle has not arrived, and the processing is finished.
  • sampling value report count value is equal to the sampling value report interval, it means that the sampling value report message sending cycle has been reached, generate a sampling value report message, and fill the sample value
  • the fields in the report message, including information such as sampling number and sampling value, are sent to the CPU to end processing.
  • the maximum and minimum average value report is also required, but the redundancy suppression function is not required.
  • the user parameters are preset to the CPU, and the redundancy flag is inhibited from being set to a non-true value.
  • the CPU calculates and obtains an unoccupied sampling number, and configures the sampling parameters to the coprocessor.
  • the sampling parameters include sampling number, sampling frequency, sampling function items, sampling value reporting interval, average reporting interval, and suppressing redundant identification. Wait.
  • the coprocessor According to the sampling parameters configured by the CPU, the coprocessor generates a sampling number, sampling function item, sampling value report count value, sampling value report interval, and average value report for the sampling function items whose sampling frequency is not 0 at the configured sampling frequency.
  • the count value, the average reporting interval, and the sampling indication message of the suppression redundancy flag are sent to the NPU.
  • the sampling value report count value can start from 1, each time a sampling indication message is sent, if the sampling value report count value corresponding to the sampling function item is not equal to the corresponding sampling value report interval, the sampling in the next sampling indication message sent The sampling value report count value corresponding to the function item is incremented by 1; if the sampling value report count value corresponding to the sampling function item is equal to the corresponding sampling value report interval, the sampling value report corresponding to the sampling function item in the next sampling indication message sent The count value is set to 1.
  • the average value report count value in the sampling indication message starts from 1.
  • the average report count value corresponding to the sampling function item is not 0
  • the average report count value corresponding to the sampling function item in the next sampling instruction message sent increases by 1; if the average report count value corresponding to the sampling function item is equal to the corresponding average report interval , The average report count value corresponding to the sampling function item in the next sampling instruction message to be sent is set to 1.
  • sampling function items are related, they have the same sampling frequency and report frequency and can be configured at the same time. Multiple sampling function items can be carried in the sampling instruction message at the same time, and the sampling function can be carried in the sampling instruction message The total number of items field.
  • the NPU After receiving the sampling instruction message, the NPU recognizes the sampling function items in the sampling instruction message, and reads the corresponding statistics, alarms, and status values.
  • the NPU judges that the suppression redundancy flag is not a true value, it moves the sample value to the matched sample value sequence according to the sampling instruction message, and updates the maximum value and the minimum value in the period; the NPU according to the sampling instruction message If the average value report interval is not zero, it is judged that the maximum and minimum average values need to be reported.
  • the NPU judges whether the average value report count value and the average value report interval in the sampling indication message are equal. If the average value report count value and the average value report interval are not equal , It means that the average value report message sending period has not arrived. If the average value report count value and the average value report interval are equal, it means that the average value report message sending period has come.
  • the fields in, including the sampling number, average value and other information, are sent to the CPU, and the corresponding maximum and minimum values in the period are cleared in the storage space; in the NPU judges the sampling indication message, the sampling value reports the count value and the sampling value Whether the report interval is equal, if the sampling value report count value and the sampling value report interval are not equal, it means that the sampling value report message transmission period has not been reached, and the processing is ended.
  • the sampling value report message sending cycle is to generate a sampling value report message, fill in the fields in the sampling value report message, including information such as sampling number and sampling value, and send it to the CPU to end processing.
  • a redundancy suppression function is also required, the user parameters are preset to the CPU, and the suppression redundancy flag is set to a true value.
  • the NPU After receiving the sampling instruction message, the NPU recognizes the sampling function items in the sampling instruction message, and reads the corresponding statistics, alarms, and status values.
  • the NPU When the NPU judges that the suppression redundancy flag is a true value, it moves the sample value to the matched sample value sequence according to the sampling instruction message, and updates the maximum value and minimum value in the period to determine whether the sample value has changed at the current moment. For statistical values, if the sampling value at the current moment is not zero, it is considered that the sampling value at the current moment has changed. For the alarm value and status value, if the sampling value at the current moment is inconsistent with the sampling value at the previous moment, it is considered as the current sampling value. The sampled value has changed at the moment.
  • sampled value report message is generated, and the fields in the sampled value report message are filled, including information such as sample number and sampled value, and sent to the CPU to end processing; if the sampled value at the current time is not If there is a change, it is judged whether the sampling value report count value and the sampling value report interval in the sampling indication message are equal.
  • sampling value report count value and the sampling value report interval are not equal, it means that the sampling value report message sending cycle has not arrived, and the end Processing, if the sampling value report count value is equal to the sampling value report interval, it means that the sampling value report message sending cycle has been reached, a sampling value report message is generated, and the fields in the sampling value report message are filled in, including sampling number and sampling value Wait for the information, send it to the CPU, and end the processing.
  • FIG. 4 is a flowchart of another data sampling method for network equipment provided by an embodiment of the application.
  • the CPU calculates and configures sampling parameters, including sampling number, sampling frequency, sampling value reporting interval, average reporting interval, and suppressing redundancy identification.
  • the CPU configures the sampling parameter to the sampling parameter configuration table of the coprocessor.
  • the coprocessor generates a sampling instruction message according to the sampling parameter configuration table, and sends it to the NPU.
  • the NPU reads the corresponding sampling value according to the sampling instruction message, and processes it.
  • the NPU judges whether the redundancy suppression flag is a true value according to the sampling instruction message, if the redundancy suppression flag is a true value, execute S2110, and if the redundancy suppression flag is not a true value, execute S260.
  • the NPU moves the sampled value to the storage space according to the sample number, and updates the maximum value in the period and the minimum value in the period.
  • S270 The NPU judges whether the average value report interval is non-zero and the average value report count value is equal to the average value report interval. If the average value report interval is non-zero and the average value report count value is equal to the average value report interval, execute S2100. If the value report interval is zero or the average value report count value is not equal to the average value report interval, S280 is executed.
  • the NPU judges whether the sample value report count value is equal to the sample value report interval. If the sample value report count value is equal to the sample value report interval, execute S290; if the sample value report count value is not equal to the sample value report interval, execute S2130.
  • the NPU generates a sample value report message, fills in the sample value, and sends it to the CPU.
  • the S2100 and NPU generate average value report messages, fill in the maximum and minimum average values, and send them to the CPU to clear the maximum and minimum values in the period in the storage space.
  • the NPU moves the current sample value to the corresponding storage space.
  • the NPU judges whether the sampling value at the current moment has changed. If the sampling value at the current moment has changed, S290 is executed, and if the sampling value at the current moment has not changed, S280 is executed.
  • the technical solution provided by this implementation mode uses the coprocessor to quickly send sampling messages to the NPU for rapid sampling.
  • the NPU moves the sampled values to the storage space, and then periodically summarizes the method of sending report messages to the CPU to achieve rapid access to the network device. Sampling of performance, alarms and status.
  • the above technical solution can achieve a higher sampling frequency (the statistical data can be sampled at a frequency faster than 100 times per second at the fastest), and it will not consume a large amount of CPU processing resources.
  • the message channel bandwidth between the CPU and NPU is saved. Furthermore, the statistical monitoring performance of the entire network data is improved, and the reliability of the telecommunication service is guaranteed.
  • FIG. 5 is a schematic flowchart of another data sampling method for a network device according to an embodiment of the application.
  • This method is suitable for sampling the business (or performance) statistics, real-time alarms, real-time status, etc. processed by the network equipment at a high frequency.
  • the method can be executed by the data sampling device applied to the network device in the CPU provided by the present application, and the data sampling device of the network device can be implemented by software and/or hardware and integrated in the CPU.
  • the data sampling method applied to the network device in the CPU provided by this application includes:
  • sampling parameter of the target data to the coprocessor, where the sampling parameter is used by the coprocessor to generate a sampling instruction message at a set frequency and send the sampling instruction message to the NPU, and The sampling instruction message is used by the NPU to generate a sampling report message of the target data and send the sampling report message to the CPU.
  • sampling parameters at least include: sampling number, sampling function item, sampling frequency, and reporting interval.
  • the reporting interval includes: sampling value reporting interval and/or average value reporting interval.
  • sampling value report interval defines the data sampling value to be reported every sampling times
  • average reporting interval defines the data average value to be reported every sampling times.
  • the sampling value report interval is determined according to the sampling value report frequency and the sampling frequency.
  • the average value report interval is determined according to the average value report frequency and the sampling frequency.
  • the average value may refer to the average value of all sampled values.
  • the average value may refer to the average value of the maximum sampling value and the minimum sampling value, that is, the maximum and minimum average values.
  • the average value is the maximum and minimum average value as an example for explanation.
  • sampling parameters include at least: sampling number, sampling function item, sampling frequency, reporting interval, and suppressing redundancy identification.
  • Redundancy suppression flag a parameter set by the user, used to identify whether the redundancy suppression function is required during the sampling process.
  • the suppression of redundancy is identified as a valid value, which corresponds to the need for redundancy suppression in the sampling process.
  • the coprocessor reads the sampling parameters of the target data configured by the CPU, generates a sampling instruction message at a set frequency according to the sampling parameters, and sends the sampling instruction message to the NPU.
  • the co-processor generating a sampling instruction message at a set frequency according to the sampling parameter may be: the co-processor, according to the sampling parameter, performs a sampling function item whose sampling frequency is not zero, using the sampling frequency Generate a sampling instruction message.
  • a logic table may be maintained inside the coprocessor, which may be referred to as a sampling parameter configuration table here, which is used to describe the sampling function items that need to be performed currently and the corresponding parameters.
  • the quick sampling parameter configuration table is a logical table that the coprocessor can access. It can be configured by the CPU, that is, the CPU can configure the sampling parameters into the sampling parameter configuration table, so as to control the coprocessor based on the sampling parameter configuration table.
  • the sampling parameter generates a sampling instruction message and sends it.
  • the sampling indication message carries a message type, sampling number, sampling function item, report count value, and reporting interval; or, the sampling indication message carries a message type, sampling number, and sampling function Item, report count value, report interval, and redundancy suppression flag; when the report interval includes the sample value report interval, the report count value includes the sample value report count value; in the sampling instruction messages sent sequentially, The sample value report count value is sequentially increased from the initial value, and reset to the initial value when it increases to the sample value report interval; when the report interval includes the average value report interval, the report count value includes the average value Report count value; for the sampling function item whose average value report interval is not zero, in the sampling instruction message sent in sequence, the average value report count value is sequentially increased from the initial value to the average value Reset to the initial value at the reporting interval.
  • the sampling instruction message may further include: a total number of sampling function items, and the total number of sampling function items is used to indicate the sampling number included in the sampling instruction message and the number of sampling function items.
  • the number of groups, the sampling frequency and reporting interval of the sampling function items in different groups are the same.
  • sampling function items are related and have the same sampling frequency and report frequency, they can be configured at the same time. Furthermore, the coprocessor can carry multiple sampling function items in a sampling instruction message, and use the sampling instruction The message carries the total number of sampling function items.
  • the NPU after receiving the message on the network, the NPU will perform statistical analysis on various messages.
  • the NPU can access and read and write a piece of memory space, which is called a sampling memory space.
  • the sampling memory space is set to store sample values, for example, in the form of a sample value storage table.
  • the NPU generates a sampling report message of the target data according to the sampling instruction message, and sends the sampling report message to the CPU.
  • the NPU generates a sampling report message of the target data according to the sampling indication message, which may be:
  • the NPU updates the locally stored sampling value according to the sampling instruction message; when the report count value in the sampling instruction message is equal to the reporting interval, generates a sample of the target data according to the locally stored sampling value Report message; wherein, the sampling report message includes a sampling value report message and/or an average value report message.
  • the NPU generates a sampling report message of the target data according to the sampling indication message, which may be:
  • the NPU updates the locally stored sampling value according to the sampling indication message, and determines whether the sampling value at the current time has changed; if the current time is If the sampling value changes, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; if the sampling value at the current moment does not change, when all the sampling indication messages are When the report count value is equal to the report interval, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; wherein, the sampling report message includes a sampling value report message and/or Average report message.
  • the CPU receives the sampling report message of the target data sent by the NPU, analyzes it, and obtains the sampling value carried in the sampling report message, and performs setting processing on the sampling value to present it to the man-machine interface.
  • the technical solution provided by this embodiment can achieve a higher sampling frequency (the statistical data can be sampled at a frequency faster than 100 times per second at the fastest) and will not consume a lot of money.
  • the CPU processing resources also save the bandwidth of the message channel between the CPU and the NPU. Furthermore, the statistical monitoring performance of the entire network data is improved, and the reliability of the telecommunication service is guaranteed.
  • FIG. 6 is a schematic flowchart of another data sampling method for a network device according to an embodiment of the application.
  • This method is suitable for sampling the business (or performance) statistics, real-time alarms, real-time status, etc. processed by the network equipment at a high frequency.
  • the method can be executed by the data sampling device applied to the network device in the coprocessor provided in this application, and the data sampling device of the network device can be implemented by software and/or hardware and integrated in the coprocessor.
  • the coprocessor is an external FPGA, or a packet generator in the NPU, which is not limited in this embodiment, and devices with simple calculation processing capabilities and fast packet sending capabilities in network equipment can all be used as co-processing Device.
  • the data sampling method applied to the network device in the coprocessor provided by this application includes:
  • sampling parameters at least include: sampling number, sampling function item, sampling frequency, and reporting interval.
  • the reporting interval includes: sampling value reporting interval and/or average value reporting interval.
  • sampling value report interval defines the data sampling value to be reported every sampling times
  • average reporting interval defines the data average value to be reported every sampling times.
  • the sampling value report interval is determined according to the sampling value report frequency and the sampling frequency.
  • the average value report interval is determined according to the average value report frequency and the sampling frequency.
  • the average value may refer to the average value of all sampled values.
  • the average value may refer to the average value of the maximum sampling value and the minimum sampling value, that is, the maximum and minimum average values.
  • the average value is the maximum and minimum average value as an example for explanation.
  • sampling parameters include at least: sampling number, sampling function item, sampling frequency, reporting interval, and suppressing redundancy identification.
  • Redundancy suppression flag a parameter set by the user, used to identify whether the redundancy suppression function is required during the sampling process.
  • the suppression of redundancy is identified as a valid value, which corresponds to the need for redundancy suppression in the sampling process.
  • the coprocessor generates a sampling instruction message at a set frequency according to the sampling parameter. Frequency generates sampling indication message.
  • a logic table may be maintained inside the coprocessor, which may be referred to as a sampling parameter configuration table here, which is used to describe the sampling function items that need to be performed currently and the corresponding parameters.
  • the quick sampling parameter configuration table is a logical table that the coprocessor can access. It can be configured by the CPU, that is, the CPU can configure the sampling parameters into the sampling parameter configuration table, so as to control the coprocessor based on the sampling parameter configuration table. Sampling parameters are used to send sampling instructions.
  • the sampling indication message carries a message type, sampling number, sampling function item, report count value, and reporting interval; or, the sampling indication message carries a message type, sampling number, and sampling function Item, report count value, report interval, and redundancy suppression flag; when the report interval includes the sample value report interval, the report count value includes the sample value report count value; in the sampling instruction messages sent sequentially, The sample value report count value is sequentially increased from the initial value, and reset to the initial value when it increases to the sample value report interval; when the report interval includes the average value report interval, the report count value includes the average value Report count value; for the sampling function item whose average value report interval is not zero, in the sampling instruction message sent in sequence, the average value report count value is sequentially increased from the initial value to the average value Reset to the initial value at the reporting interval.
  • the sampling instruction message may further include: a total number of sampling function items, and the total number of sampling function items is used to indicate the sampling number included in the sampling instruction message and the number of sampling function items.
  • the number of groups, the sampling frequency and reporting interval of the sampling function items in different groups are the same.
  • sampling function items are related and have the same sampling frequency and report frequency, they can be configured at the same time. Furthermore, the coprocessor can carry multiple sampling function items in a sampling instruction message, and use the sampling instruction The message carries the total number of sampling function items.
  • the NPU after receiving the message on the network, the NPU will perform statistical analysis on various messages.
  • the NPU can access and read and write a piece of memory space, which is called a sampling memory space.
  • the sampling memory space is set to store sample values, for example, in the form of a sample value storage table.
  • the NPU generates a sampling report message of the target data according to the sampling instruction message, and sends the sampling report message to the CPU.
  • the NPU generates a sampling report message of the target data according to the sampling indication message, which may be:
  • the NPU updates the locally stored sampling value according to the sampling instruction message; when the report count value in the sampling instruction message is equal to the reporting interval, generates a sample of the target data according to the locally stored sampling value Report message; wherein, the sampling report message includes a sampling value report message and/or an average value report message.
  • the NPU generates a sampling report message of the target data according to the sampling indication message, which may be:
  • the NPU updates the locally stored sampling value according to the sampling indication message, and determines whether the sampling value at the current time has changed; if the current time is If the sampling value changes, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; if the sampling value at the current moment does not change, when all the sampling indication messages are When the report count value is equal to the report interval, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; wherein, the sampling report message includes a sampling value report message and/or Average report message.
  • the CPU receives the sampling report message of the target data sent by the NPU, analyzes it, and obtains the sampling value carried in the sampling report message, and performs setting processing on the sampling value to present it to the man-machine interface.
  • the technical solution provided by this embodiment can achieve a higher sampling frequency (the statistical data can be sampled at a frequency faster than 100 times per second at the fastest) and will not consume a lot of money.
  • the CPU processing resources also save the bandwidth of the message channel between the CPU and the NPU. Furthermore, the statistical monitoring performance of the entire network data is improved, and the reliability of the telecommunication service is guaranteed.
  • FIG. 7 is a schematic flowchart of another data sampling method for a network device according to an embodiment of the application.
  • This method is suitable for sampling the business (or performance) statistics, real-time alarms, real-time status, etc. processed by the network equipment at a high frequency.
  • the method can be executed by the data sampling device applied to the network device in the NPU provided in this application, and the data sampling device of the network device can be implemented by software and/or hardware and integrated in the NPU.
  • the data sampling method applied to the network device in the NPU provided by this application includes:
  • S510 Receive a sampling instruction message sent by the coprocessor, where the sampling instruction message is generated and sent by the coprocessor at a set frequency according to the sampling parameter of the target data sent by the CPU.
  • the CPU configures the sampling parameters of the target data to the coprocessor.
  • the sampling parameters at least include: sampling number, sampling function item, sampling frequency, and reporting interval.
  • the reporting interval includes: sampling value reporting interval and/or average value reporting interval.
  • sampling value report interval defines the data sampling value to be reported every sampling times
  • average reporting interval defines the data average value to be reported every sampling times.
  • the sampling value report interval is determined according to the sampling value report frequency and the sampling frequency.
  • the average value report interval is determined according to the average value report frequency and the sampling frequency.
  • the average value may refer to the average value of all sampled values.
  • the average value may refer to the average value of the maximum sampling value and the minimum sampling value, that is, the maximum and minimum average values.
  • the average value is the maximum and minimum average value as an example for explanation.
  • sampling parameters include at least: sampling number, sampling function item, sampling frequency, reporting interval, and suppressing redundancy identification.
  • Redundancy suppression flag a parameter set by the user, used to identify whether the redundancy suppression function is required during the sampling process.
  • the suppression of redundancy is identified as a valid value, which corresponds to the need for redundancy suppression in the sampling process.
  • the coprocessor generates a sampling instruction message at a set frequency according to the sampling parameter and sends it to the NPU.
  • the coprocessor generates a sampling instruction message at a set frequency according to the sampling parameter. Frequency generates sampling indication message.
  • the coprocessor internally maintains a logical table, which can be called a sampling parameter configuration table, which is used to describe the current sampling function items and corresponding parameters that need to be performed.
  • the quick sampling parameter configuration table is a logical table that the coprocessor can access. It can be configured by the CPU, that is, the CPU can configure the sampling parameters into the sampling parameter configuration table, so as to control the coprocessor based on the sampling parameter configuration table. Sampling parameters are used to send sampling instructions.
  • the sampling indication message carries a message type, sampling number, sampling function item, report count value, and reporting interval; or, the sampling indication message carries a message type, sampling number, and sampling function Item, report count value, report interval, and redundancy suppression flag; when the report interval includes the sample value report interval, the report count value includes the sample value report count value; in the sampling instruction messages sent sequentially, The sample value report count value is sequentially increased from the initial value, and reset to the initial value when it increases to the sample value report interval; when the report interval includes the average value report interval, the report count value includes the average value Report count value; for the sampling function item whose average value report interval is not zero, in the sampling instruction message sent in sequence, the average value report count value is sequentially increased from the initial value to the average value Reset to the initial value at the reporting interval.
  • the sampling instruction message may further include: a total number of sampling function items, and the total number of sampling function items is used to indicate the sampling number included in the sampling instruction message and the number of sampling function items.
  • the number of groups, the sampling frequency and reporting interval of the sampling function items in different groups are the same.
  • sampling function items are related and have the same sampling frequency and report frequency, they can be configured at the same time. Furthermore, the coprocessor can carry multiple sampling function items in a sampling instruction message, and use the sampling instruction The message carries the total number of sampling function items.
  • S520 Generate a sampling report message of the target data according to the sampling instruction message, and send the sampling report message to the CPU.
  • the NPU after receiving the message on the network, the NPU will perform statistical analysis on various messages.
  • the NPU can access and read and write a piece of memory space, which is called a sampling memory space.
  • the sampling memory space is set to store sample values, for example, in the form of a sample value storage table.
  • the NPU generates a sampling report message of the target data according to the sampling indication message, which may be:
  • the NPU generates a sampling report message of the target data according to the sampling indication message, which may be:
  • the NPU updates the locally stored sampling value according to the sampling indication message, and determines whether the sampling value at the current time has changed; if the current time is If the sampling value changes, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; if the sampling value at the current moment does not change, when all the sampling indication messages are When the report count value is equal to the report interval, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; wherein, the sampling report message includes a sampling value report message and/or Average report message.
  • the CPU receives the sampling report message of the target data sent by the NPU, analyzes it, and obtains the sampling value carried in the sampling report message, and performs setting processing on the sampling value to present it to the man-machine interface.
  • the technical solution provided by this embodiment can achieve a higher sampling frequency (the statistical data can be sampled at a frequency faster than 100 times per second at the fastest) and will not consume a lot of money.
  • the CPU processing resources also save the bandwidth of the message channel between the CPU and the NPU. Furthermore, the statistical monitoring performance of the entire network data is improved, and the reliability of the telecommunication service is guaranteed.
  • FIG. 8 is a schematic structural diagram of a data sampling device for network equipment provided in an embodiment of the application. As shown in FIG. 8, the data sampling device applied to the network equipment in the CPU provided by the embodiment of the present application can be implemented by software and/or hardware and integrated in the CPU.
  • the device includes: a sampling parameter configuration module 610 and a sampling Report message receiving module 620.
  • the sampling parameter configuration module 610 is configured to configure sampling parameters of the target data to the coprocessor, where the sampling parameters are used by the coprocessor to generate a sampling instruction message at a set frequency and to use the sampling instruction message Sent to the NPU, the sampling instruction message is used by the NPU to generate a sampling report message of the target data and send the sampling report message to the CPU; the sampling report message receiving module 620 is configured to receive the Sampling report message of the target data sent by the NPU.
  • the technical solution provided by this embodiment can achieve a higher sampling frequency (the statistical data can be sampled at a frequency faster than 100 times per second at the fastest), and does not consume a lot of money.
  • CPU processing resources also save the bandwidth of the message channel between the CPU and the NPU.
  • the statistical monitoring performance of the entire network data is improved, and the reliability of the telecommunication service is guaranteed.
  • the sampling parameters include at least: sampling number, sampling function items, sampling frequency, and reporting interval; or, the sampling parameters include at least sampling number, sampling function items, sampling frequency, reporting interval, and suppressing redundancy.
  • Identification wherein, the reporting interval includes: sampling value reporting interval and/or average value reporting interval, the sampling value reporting interval is determined according to the sampling value reporting frequency and the sampling frequency, the average reporting interval is based on the average value reporting The frequency and the sampling frequency are determined.
  • the sampling indication message carries a message type, sampling number, sampling function item, report count value, and reporting interval; or, the sampling indication message carries a message type, sampling number, and sampling function Item, report count value, report interval, and redundancy suppression flag; when the report interval includes the sample value report interval, the report count value includes the sample value report count value; in the sampling instruction messages sent sequentially, The sample value report count value is sequentially increased from the initial value, and reset to the initial value when it increases to the sample value report interval; when the report interval includes the average value report interval, the report count value includes the average value Report count value; for the sampling function item whose average value report interval is not zero, in the sampling instruction message sent in sequence, the average value report count value is sequentially increased from the initial value to the average value Reset to the initial value at the reporting interval.
  • the sampling instruction message further includes: a total number of sampling function items, and the total number of sampling function items is used to indicate the sampling number and the number of groups of sampling function items included in the sampling instruction message, The sampling frequency and reporting interval of the sampling function items in different groups are the same.
  • the data sampling device applied to the network equipment in the CPU provided in this embodiment is configured to implement the data sampling method applied to the network equipment in the CPU as described in the embodiment of this application.
  • the network application in the CPU provided in this embodiment is The implementation principle and technical effect of the data sampling device of the device are similar to the data sampling method applied to the network device in the CPU described in the embodiment of the present application, and will not be repeated here.
  • FIG. 9 is a schematic structural diagram of another data sampling device for a network device according to an embodiment of the application.
  • the data sampling device applied to the network device in the coprocessor provided by the embodiment of the present application can be implemented by software and/or hardware and integrated in the coprocessor.
  • the device includes: sampling parameter reading The fetching module 710 and the sampling instruction message sending module 720.
  • the sampling parameter reading module 710 is configured to read the sampling parameters of the target data configured by the CPU; the sampling instruction message sending module 720 is configured to generate a sampling instruction message at a set frequency according to the sampling parameters, and to combine the sampling An instruction message is sent to the NPU, and the sampling instruction message is used by the NPU to generate a sampling report message of the target data and send the sampling report message to the CPU.
  • the technical solution provided by this embodiment can achieve a higher sampling frequency (the statistical data can be sampled at a frequency faster than 100 times per second at the fastest), and does not consume a lot of money.
  • CPU processing resources also save the bandwidth of the message channel between the CPU and the NPU.
  • the statistical monitoring performance of the entire network data is improved, and the reliability of the telecommunication service is guaranteed.
  • the sampling parameters include at least: sampling number, sampling function items, sampling frequency, and reporting interval; or, the sampling parameters include at least sampling number, sampling function items, sampling frequency, reporting interval, and suppressing redundancy.
  • Identification wherein, the reporting interval includes: sampling value reporting interval and/or average value reporting interval, the sampling value reporting interval is determined according to the sampling value reporting frequency and the sampling frequency, the average reporting interval is based on the average value reporting The frequency and the sampling frequency are determined.
  • the sampling instruction message sending module 720 is configured to generate a sampling instruction message at the sampling frequency for a sampling function item whose sampling frequency is not zero according to the sampling parameter, and send the sampling instruction message
  • the message is sent to the NPU; the sampling instruction message carries the message type, sampling number, sampling function item, report count value, and reporting interval; or, the sampling instruction message carries the message type, sampling number, and sampling function Item, report count value, report interval, and redundancy suppression flag; when the report interval includes the sample value report interval, the report count value includes the sample value report count value; in the sampling instruction messages sent sequentially, The sample value report count value is sequentially increased from the initial value, and reset to the initial value when it increases to the sample value report interval; when the report interval includes the average value report interval, the report count value includes the average value Report count value; for the sampling function item whose average value report interval is not zero, in the sampling instruction message sent in sequence, the average value report count value is sequentially increased from the initial value to the average value Reset to the initial value at the
  • the sampling instruction message further includes: a total number of sampling function items, where the total number of sampling function items is used to indicate the sampling number and the group number of sampling function items included in the sampling instruction message, The sampling frequency and reporting interval of the sampling function items in different groups are the same.
  • the data sampling device applied to the network device in the coprocessor provided in this embodiment is configured to implement the data sampling method applied to the network device in the coprocessor as described in the embodiment of this application.
  • the implementation principle and technical effect of the data sampling device of the network device in the coprocessor are similar to the data sampling method applied to the network device in the coprocessor described in the embodiment of the present application, and will not be repeated here.
  • FIG. 10 is a schematic structural diagram of another data sampling device for a network device provided in an embodiment of the application.
  • the data sampling device applied to the network equipment in the NPU provided by the embodiment of the present application can be implemented by software and/or hardware and integrated in the NPU.
  • the device includes: a sampling instruction message receiving module 810 And the sampling report message sending module 820.
  • the sampling instruction message receiving module 810 is configured to receive a sampling instruction message sent by the coprocessor.
  • the sampling instruction message is generated and sent by the coprocessor at a set frequency according to the sampling parameters of the target data sent by the CPU.
  • the sampling report message sending module 820 is configured to generate a sampling report message of the target data according to the sampling instruction message, and send the sampling report message to the CPU.
  • the technical solution provided by this embodiment can achieve a higher sampling frequency (the statistical data can be sampled at a frequency faster than 100 times per second at the fastest), and does not consume a lot of money.
  • CPU processing resources also save the bandwidth of the message channel between the CPU and the NPU.
  • the statistical monitoring performance of the entire network data is improved, and the reliability of the telecommunication service is guaranteed.
  • the sampling parameters include at least: sampling number, sampling function items, sampling frequency, and reporting interval; or, the sampling parameters include at least sampling number, sampling function items, sampling frequency, reporting interval, and suppressing redundancy.
  • Identification wherein, the reporting interval includes: sampling value reporting interval and/or average value reporting interval, the sampling value reporting interval is determined according to the sampling value reporting frequency and the sampling frequency, the average reporting interval is based on the average value reporting The frequency and the sampling frequency are determined.
  • the sampling indication message carries a message type, sampling number, sampling function item, report count value, and reporting interval; or, the sampling indication message carries a message type, sampling number, and sampling function Item, report count value, report interval, and redundancy suppression flag; when the report interval includes the sample value report interval, the report count value includes the sample value report count value; in the sampling instruction messages sent sequentially, The sample value report count value is sequentially increased from the initial value, and reset to the initial value when it increases to the sample value report interval; when the report interval includes the average value report interval, the report count value includes the average value Report count value; for the sampling function item whose average value report interval is not zero, in the sampling instruction message sent in sequence, the average value report count value is sequentially increased from the initial value to the average value Reset to the initial value at the reporting interval.
  • the sampling report message sending module 820 is configured to update the locally stored sampling value according to the sampling indication message; when the report count value in the sampling indication message is equal to the reporting interval, according to the local storage A sampling report message of the target data is generated from the sampling value of, and sent to the CPU; wherein, the sampling report message includes a sampling value report message and/or an average value report message.
  • the sampling report message sending module 820 is configured to update the locally stored sampling value according to the sampling indication message when the suppression redundancy identifier in the sampling indication message is a valid value, and determine Whether the sampling value at the current time has changed; if the sampling value at the current time has changed, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; if the sampling value at the current time has not occurred Change, when the report count value in the sampling instruction message is equal to the reporting interval, a sampling report message of the target data is generated according to the locally stored sampling value and sent to the CPU; wherein, the The sampling report message includes a sampling value report message and/or an average value report message.
  • the data sampling device applied to the network equipment in the NPU provided in this embodiment is configured to implement the data sampling method applied to the network equipment in the NPU as described in the embodiment of this application, and the network applied to the NPU provided in this embodiment
  • the implementation principle and technical effect of the data sampling device of the device are similar to the data sampling method applied to the network device in the NPU described in the embodiment of the present application, and will not be repeated here.
  • FIG. 11 is a schematic structural diagram of a CPU provided by an embodiment of the present application.
  • the CPU provided by the present application includes: one or more processing units 910 and a storage unit 920; the processing unit 910 of the CPU may be one or more, and one processing unit 910 is taken as an example in FIG. 11;
  • the storage unit 920 is configured to store one or more programs; the one or more programs are executed by the one or more processing units 910, so that the one or more processing units 910 implements as described in the embodiments of the present application
  • the data sampling method applied to the network equipment in the CPU is configured to store one or more programs; the one or more programs are executed by the one or more processing units 910, so that the one or more processing units 910 implements as described in the embodiments of the present application.
  • the processing unit 910 and the storage unit 920 in the CPU may be connected through a bus or other methods.
  • the connection through a bus is taken as an example.
  • the storage unit 920 can be configured to store software programs, computer-executable programs, and modules, such as the program instructions/modules corresponding to the data sampling method applied to the network device in the CPU as described in the embodiments of the present application (For example, the sampling parameter configuration module 610 and the sampling report message receiving module 620 in the data sampling device applied to the network device in the CPU shown in FIG. 8).
  • the storage unit 920 may include a storage program area and a storage data area, where the storage program area may store an operating system and an application program required by at least one function; the storage data area may store data created according to the use of the device, and the like.
  • the storage unit 920 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.
  • FIG. 12 is a schematic structural diagram of a coprocessor provided by an embodiment of the application.
  • the coprocessor provided by the present application includes: one or more processing units 1010 and a storage unit 1020; the processing unit 1010 of the coprocessor may be one or more.
  • one processing unit 1010 is an example; the storage unit 1020 is configured to store one or more programs; the one or more programs are executed by the one or more processing units 1010, so that the one or more processing units 1010 implement the implementation of this application The data sampling method applied to the network device in the coprocessor described in the example.
  • the processing unit 1010 and the storage unit 1020 in the coprocessor may be connected through a bus or other methods.
  • the connection through a bus is taken as an example.
  • the storage unit 1020 can be configured to store software programs, computer-executable programs, and modules, such as the program instructions corresponding to the data sampling method applied to the network device in the coprocessor as described in the embodiment of the present application /Module (for example, the sampling parameter reading module 710 and the sampling instruction message sending module 720 in the data sampling device applied to the network device in the coprocessor shown in FIG. 9).
  • the storage unit 1020 may include a storage program area and a storage data area, where the storage program area may store an operating system and an application program required by at least one function; the storage data area may store data created according to the use of the device, and the like.
  • the storage unit 1020 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.
  • FIG. 13 is a schematic structural diagram of a coprocessor provided by an embodiment of the present application.
  • the NPU provided by this application includes: one or more processing units 1110 and a storage unit 1120; the processing unit 1110 of the NPU may be one or more, and one processing unit 1110 is taken as an example in FIG. 13;
  • the storage unit 1120 is configured to store one or more programs; the one or more programs are executed by the one or more processing units 1110, so that the one or more processing units 1110 implement The data sampling method applied to the network equipment in the NPU.
  • the processing unit 1110 and the storage unit 1120 in the coprocessor may be connected by a bus or other methods.
  • the connection by a bus is taken as an example.
  • the storage unit 1120 can be configured to store software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the data sampling method applied to the network device in the NPU as described in the embodiments of the present application (For example, the sampling instruction message receiving module 810 and the sampling report message sending module 820 in the data sampling device applied to the network device in the NPU shown in FIG. 10).
  • the storage unit 1120 may include a storage program area and a storage data area, where the storage program area may store an operating system and an application program required by at least one function; the storage data area may store data created according to the use of the device, and the like.
  • the storage unit 1120 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.
  • a network device includes: a CPU, a coprocessor, and an NPU.
  • the CPU executes the data sampling method applied to the network device in the CPU as described in any embodiment of the present application
  • the coprocessor executes the data sampling method applied to the network device in the coprocessor as described in any embodiment of the present application Data sampling method
  • the NPU executes the data sampling method applied to the network device in the NPU as described in any embodiment of this application.
  • the coprocessor is an FPGA external to the network device.
  • the coprocessor is a packet generator in the NPU.
  • the embodiment of the present application also provides a storage medium storing a computer program that, when executed by a processor, implements the data sampling method applied to the network device in the CPU in the embodiment of the present application, Either implement the data sampling method applied to the network device in the coprocessor described in the embodiment of the application, or implement the data sampling method applied to the network device in the NPU in the embodiment of the application, or implement the present application The data sampling method applied to the network device in the network device described in the embodiment.
  • the data sampling method applied to the network equipment in the CPU includes:
  • the sampling parameter of the target data is configured to the coprocessor, where the sampling parameter is used by the coprocessor to generate a sampling instruction message at a set frequency and send the sampling instruction message to the NPU, and the sampling instruction The message is used by the NPU to generate a sampling report message of the target data and send the sampling report message to the CPU; and receive a sampling report message of the target data sent by the NPU.
  • the data sampling method applied to the network equipment in the coprocessor includes:
  • the sampling report message of the target data is sent and the sampling report message is sent to the CPU.
  • the data sampling method applied to the network equipment in the NPU includes:
  • the data sampling method applied to the network equipment in the network equipment includes:
  • the sampling parameter of the target data is configured to the coprocessor through the CPU; the sampling instruction message is generated by the coprocessor at a set frequency according to the sampling parameter, and the sampling instruction message is sent to the NPU; The sampling instruction message generates a sampling report message of the target data, and sends the sampling report message to the CPU.
  • the computer-executable instructions when executed by a computer processor, they can also be used to implement any of the data sampling methods applied to the network device in the CPU in the embodiments of the present application, or to implement the Any of the data sampling methods applied to the network device in the coprocessor, or the data sampling method applied to the network device in the NPU in the embodiments of this application, or the implementation of the data sampling method in the embodiments of this application Any of the aforementioned data sampling methods applied to network equipment in network equipment.
  • this application can be implemented by software and necessary general-purpose hardware, or can be implemented by hardware.
  • the technical solution of this application can essentially be embodied in the form of a software product.
  • Software products can be stored in computer readable storage media, such as computer floppy disks, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc. , Including multiple instructions to enable a communication device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments of the present application.
  • a communication device which may be a personal computer, a server, or a network device, etc.
  • the multiple units and modules included are only divided according to the functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be realized; in addition, the names of multiple functional units are only for the convenience of distinguishing each other. , Is not used to limit the scope of protection of this application.
  • the various embodiments of the present application can be implemented in hardware or dedicated circuits, software, logic or any combination thereof.
  • some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device, although the present application is not limited thereto.
  • Computer program instructions can be assembly instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or written in any combination of one or more programming languages Source code or object code.
  • ISA Instruction Set Architecture
  • the block diagram of any logic flow in the drawings of the present application may represent program steps, or may represent interconnected logic circuits, modules, and functions, or may represent a combination of program steps and logic circuits, modules, and functions.
  • the computer program can be stored on the memory.
  • the memory can be of any type suitable for the local technical environment and can be implemented using any suitable data storage technology, such as but not limited to read only memory (ROM), random access memory (RAM), optical storage devices and systems (digital multi-function optical discs) (Digital Video Disc, DVD) or Compact Disk (CD)), etc.
  • Computer-readable media may include non-transitory storage media.
  • the data processor can be any type suitable for the local technical environment, such as but not limited to general-purpose computers, special-purpose computers, microprocessors, digital signal processors (Digital Signal Processing, DSP), application specific integrated circuits (ASICs) ), FPGA and processors based on multi-core processor architecture.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Computer Hardware Design (AREA)
  • Mathematical Physics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Debugging And Monitoring (AREA)
  • Maintenance And Management Of Digital Transmission (AREA)

Abstract

本文公开一种网络设备的数据采样方法、装置、设备及介质。该网络设备的数据采样方法包括：通过CPU将目标数据的采样参数配置至协处理器；通过所述协处理器根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至NPU；通过所述NPU根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至所述CPU。

Description

网络设备的数据采样方法、装置、设备及介质 技术领域
本申请涉及通信技术领域，例如涉及一种网络设备的数据采样方法、装置、设备及介质。
背景技术
随着第五代移动通信系统(the 5th Generation mobile communication system，5G)时代的到来，运营商配置管理接口的兴起，数据流量呈指数增长，网络所承载的业务也逐渐趋向于多元化，这对电信传输设备的可维护性和监控性能提出了更高的要求。其中，网络设备的性能、告警、状态等的统计采样会达到每秒最高100次。
传统的报文统计采样方式是：在报文经过网络设备时，由网络设备的网络处理器(Network Processing Unit，NPU)进行报文统计，并由NPU将统计的多项统计值存储在NPU的存储器中，再由设备的中央处理单元(Central Processing Unit，CPU)定时向NPU发送查询消息，定时采样获取的统计值(如图1所示)，并在处理后呈现到人机界面。但是，在查询项目过多时CPU需要处理成倍的查询消息，严重消耗了CPU的处理资源，以及CPU到NPU之间的配置查询通道带宽。针对高达10ms-100ms的采样要求，CPU在短时间内无法处理如此多的消息，即使有高性能CPU能够处理如此多的消息也会消耗高性能CPU的大量的处理资源，而且高性能CPU的价格十分昂贵。
发明内容
本申请提供一种网络设备的数据采样方法、装置、设备及介质，以在满足网络设备数据高采样率要求的前提下，极大地减少占用的CPU处理资源。
提供一种网络设备的数据采样方法，应用于网络设备，包括：
通过CPU将目标数据的采样参数配置至协处理器；通过所述协处理器根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至NPU；通过NPU根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至所述CPU。
还提供一种网络设备的数据采样方法，应用于CPU，包括：
将目标数据的采样参数配置至协处理器，其中，所述采样参数用于所述协 处理器以设定频率生成采样指示报文并将所述采样指示报文发送至NPU，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至CPU；接收所述NPU发送的所述目标数据的采样报告报文。
还提供一种网络设备的数据采样方法，应用于协处理器，包括：
读取CPU配置的目标数据的采样参数；根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至NPU，其中，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至CPU。
还提供一种网络设备的数据采样方法，应用于NPU，包括：
接收协处理器发送的采样指示报文，其中，所述采样指示报文为所述协处理器根据CPU发送的目标数据的采样参数以设定频率生成并发送的；根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至CPU。
还提供一种网络设备的数据采样装置，应用于CPU，包括：
采样参数配置模块，设置为将目标数据的采样参数配置至协处理器，其中，所述采样参数用于所述协处理器以设定频率生成采样指示报文并将所述采样指示报文发送至NPU，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至CPU；采样报告报文接收模块，设置为接收所述NPU发送的所述目标数据的采样报告报文。
还提供一种网络设备的数据采样装置，应用于协处理器，包括：
采样参数读取模块，设置为读取CPU配置的目标数据的采样参数；采样指示报文发送模块，设置为根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至NPU，其中，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至CPU。
还提供一种网络设备的数据采样装置，应用于NPU，包括：
采样指示报文接收模块，设置为接收协处理器发送的采样指示报文，其中，所述采样指示报文为所述协处理器根据CPU发送的目标数据的采样参数以设定频率生成并发送的；采样报告报文发送模块，设置为根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至CPU。
还提供一种CPU，包括：一个或多个处理单元；存储单元，用于存储一个或多个程序；当所述一个或多个程序被所述一个或多个处理单元执行，使得所述一个或多个处理单元实现上述的应用于CPU中的网络设备的数据采样方法。
还提供一种协处理器，包括：一个或多个处理单元；存储单元，用于存储一个或多个程序；当所述一个或多个程序被所述一个或多个处理单元执行，使得所述一个或多个处理单元实现上述的应用于协处理器中的网络设备的数据采样方法。
还提供一种NPU，包括：一个或多个处理单元；存储单元，用于存储一个或多个程序；当所述一个或多个程序被所述一个或多个处理单元执行，使得所述一个或多个处理单元实现上述的应用于NPU中的网络设备的数据采样方法。
还提供一种网络设备，包括：CPU、协处理器和NPU，其中，所述CPU执行上述的应用于CPU中的网络设备的数据采样方法，所述协处理器执行上述的应用于协处理器中的网络设备的数据采样方法，所述NPU执行上述的应用于NPU中的网络设备的数据采样方法。
还提供了一种存储介质，所述存储介质存储有计算机程序，所述计算机程序被处理器执行时实现上述的网络设备的数据采样方法。
附图说明
图1为传统的网络设备的模块结构示意图；
图2为本申请实施例提供的一种网络设备的模块结构示意图；
图3为本申请实施例提供的一种网络设备的数据采样方法的流程示意图；
图4为本申请实施例提供的另一种网络设备的数据采样方法的流程示意图；
图5为本申请实施例提供的另一种网络设备的数据采样方法的流程示意图；
图6为本申请实施例提供的另一种网络设备的数据采样方法的流程示意图；
图7为本申请实施例提供的另一种网络设备的数据采样方法的流程示意图；
图8为本申请实施例提供的一种网络设备的数据采样装置的模块结构示意图；
图9为本申请实施例提供的另一种网络设备的数据采样装置的模块结构示意图；
图10为本申请实施例提供的另一种网络设备的数据采样装置的模块结构示意图；
图11为本申请实施例提供的一种CPU的结构示意图；
图12为本申请实施例提供的一种协处理器的结构示意图；
图13为本申请实施例提供的一种NPU的结构示意图。
具体实施方式
下文中将结合附图对本申请的实施例进行说明。
如图2所示，本申请提供的网络设备包括：CPU、协处理器和NPU。
CPU，设置为处理用户配置和采样结果处理、显示，在系统中承担总体控制作用；NPU，是网络设备的核心器件，具有一定的编程能力，能对网络报文进行高速处理，在本申请中，NPU接收网络上的报文之后会对报文进行多种统计。协处理器，可以是外挂现场可编程逻辑门阵列(Field Programmable Gate Array，FPGA)或者NPU内部的包发生器，其具有简单的计算处理能力和快速发包能力。
CPU将目标数据的采样参数配置至协处理器，协处理器接收到所述采样参数之后，根据所述采样参数以设定频率生成采样指示报文发送至NPU，NPU根据所述采样指示报文生成目标数据的采样报告报文，发送至CPU，以使CPU在接收到采样报告报文之后，对采样报告报文中携带的采样结果进行处理。
在一个示例性实施方式中，图3为本申请实施例提供的一种网络设备的数据采样方法的流程示意图。该方法适用于以高频率对网络设备处理的业务(或性能)统计、实时告警、实时状态等进行采样的情况。该方法可以由本申请提供的应用于网络设备中的网络设备的数据采样装置执行，该网络设备的数据采样装置可以由软件和/或硬件实现，并集成在网络设备中。
如图3所示，本申请提供的应用于网络设备中的网络设备的数据采样方法，包括：
S110、通过CPU将目标数据的采样参数配置至协处理器。
目标数据，指的是需要对网络设备进行采样而得到的数据，例如是设备处理的业务统计数据、实时告警数据、实时状态数据等。本实施例所述的数据采样方法，涉及到网络设备端口报文收发、多层次业务、访问控制列表(Access Control List，ACL)等统计，以及多种告警、多种状态的采样。
首先将用户配置参数预设到CPU，CPU将对用户配置参数进行处理后，得到采样参数，配置至协处理器。可选的，协处理器为外挂的FPGA，或者为NPU中的包发生器，本实施例对此不做限定，网络设备中具有简单的计算处理能力和快速发包能力的装置均可以作为协处理器。
用户配置参数，可以是采样功能项、采样频率、报告频率等。报告频率可以是采样值报告频率和/或平均值报告频率。用户配置参数，还可以包括是否抑 制冗余。
在一示例中，平均值可以指周期内所有采样值的平均值。
在另一示例中，平均值可以指周期内最大采样值和周期内最小采样值的平均值，即最大最小平均值。
在一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔。
采样编号，用于唯一标识采样任务；采样功能项，用于标识采样任务的名称，例如是业务统计1、业务统计2、告警1、告警2、状态1、状态2等；采样频率为用户设置的参数，定义了每秒从连续的数据中取出的数据的数量，也称采样速率；报告间隔，定义了每采样多少次就上报一次。
采样值报告间隔根据报告频率和采样频率确定，其中，报告频率为用户设置的参数，定义了每秒上报数据的次数。例如，报告间隔＝采样频率/报告频率。例如，报告频率为10次/秒，采样频率为100次/秒，则报告间隔为10，即为每采样十次报告一次。一实施例中，采样频率为报告频率的整数倍。
所述报告间隔包括：采样值报告间隔和/或平均值报告间隔。其中，采样值报告间隔，定义了每采样多少次就上报一次数据采样值；平均值报告间隔，定义了每采样多少次就上报一次数据平均值。
采样值报告间隔根据采样值报告频率和所述采样频率确定，其中，采样值报告频率为用户设置的参数，定义了每秒上报数据采样值的次数，例如，采样值报告间隔＝采样频率/采样值报告频率。
平均值报告间隔根据平均值报告频率和所述采样频率确定，其中，平均值报告频率为用户设置的参数，定义了每秒上报数据平均值的次数，例如，平均值报告间隔＝采样频率/平均值报告频率。
将用户配置参数预设到CPU之后，CPU根据采样频率和报告频率计算出报告间隔。
在另一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔、抑制冗余标识。
抑制冗余标识，为用户设置的参数，用于标识采样过程中是否需要冗余抑制功能。抑制冗余标识为有效值，对应于采样过程中需要冗余抑制功能。
CPU向协处理器配置的采样参数，可以是至少一组。也即CPU可以针对多个采样功能项向协处理器配置对应的采样参数。
S120、通过所述协处理器根据所述采样参数以设定频率生成采样指示报文， 并将所述采样指示报文发送至NPU。
协处理器读取到CPU配置的采样参数之后，根据所述采样参数以设定频率生成采样指示报文发送至NPU。其中，设定频率可以是预设的任意频率，例如可以是协处理器的运行频率等。
在一示例中，通过所述协处理器根据所述采样参数以设定频率生成采样指示报文，可以为：通过所述协处理器根据所述采样参数，对采样频率不为零的采样功能项，以所述采样频率生成采样指示报文。
所述设定频率可以为采样参数中的采样频率。也即，协处理器以采样频率生成采样指示报文，并发送至NPU，以使NPU根据采样指示报文去获取对应的采样值。
在一示例中，协处理器内部可以维护一张逻辑表格，此处可以称之为采样参数配置表，用于表述当前需要进行的采样功能项以及对应的参数。快速采样参数配置表，是协处理器能够访问的逻辑表格，可以由CPU进行配置，也即CPU可以将采样参数配置到该采样参数配置表中，从而控制协处理器基于采样参数配置表中的采样参数进行采样指示报文的发送。一实施例中，快速采样参数配置表可以如表1所示。
在一示例中，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔。
本申请中，采样指示报文中携带的报文类型为采样类型。
报告计数值，指的是针对当次报告已发送的采样指示报文的数量，其变化范围为初始值～报告间隔。报告计数值，由初始值开始(例如是从1开始)，如果报告计数值不等于报告间隔，则协处理器发送的下一个采样指示报文中的报告计数值递增1，如果报告计数值等于报告间隔，则协处理器发送的下一个采样指示报文中的报告计数值重新设置为初始值(即为1)。
表1
Figure PCTCN2021082475-appb-000001
当所述报告间隔为采样值报告间隔时，所述报告计数值为采样值报告计数 值，在依次发送的所述采样指示报文中，所述采样值报告计数值由初始值依次递增，增至所述采样值报告间隔时重置为所述初始值。
采样值报告计数值从初始值开始(例如是从1开始)，如果采样值报告计数值不等于采样值报告间隔，则协处理器发送的下一个采样指示报文中的采样值报告计数值递增1，如果采样值报告计数值等于采样值报告间隔，则协处理器发送的下一个采样指示报文中的采样值报告计数值重新设置为初始值(即为1)。
当所述报告间隔为平均值报告间隔时，所述报告计数值为平均值报告计数值；对所述平均值报告间隔不为零的采样功能项，在依次发送的所述采样指示报文中，所述平均值报告计数值由初始值依次递增，增至所述平均值报告间隔时重置为所述初始值。
如果采样功能项对应的平均值报告间隔不为0(平均值报告间隔为0表示不需要进行平均值报告)，则采样指示报文中的平均值报告计数值从初始值开始(例如是从1开始)，如果平均值报告计数值不等于平均值报告间隔，则协处理器发送的下一个采样指示报文中的平均值报告计数值递增1，如果平均值报告计数值等于平均值报告间隔，则协处理器发送的下一个采样指示报文中的平均值报告计数值重新设置为初始值(即为1)。
设置报告计数值的意义在于，NPU不具备计时计数功能，不知道什么时候向CPU发送采样报告报文，通过采样指示报文中的报告计数值的设置，NPU可以通过比较采样指示报文中的报告计数值和报告间隔，确定出是否需要向CPU发送采样报告报文。
在另一示例中，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔、抑制冗余标识。
抑制冗余标识，用于指示NPU在抑制冗余标识为有效值时，实现冗余抑制功能。NPU在实现冗余抑制功能时，如果当次采样数据发生变化，则立即向CPU上报，如果当次采样数据没有发生变化，则按照报告间隔向CPU上报。
在一示例中，所述采样指示报文中还包括：采样功能项总数量，所述采样功能项总数量用于指示所述采样指示报文中包括的采样编号以及采样功能项的组数量；不同组中的采样功能项的采样频率和报告间隔均相同。
如果多个采样功能项是关联的，具有相同的采样频率和报告频率，则可以同时进行配置，进而，协处理器可以在一个采样指示报文中携带多个采样功能项，并在该采样指示报文中携带采样功能项总数量。
所述采样指示报文中包括的采样编号以及采样功能项的组数量为至少一组。当采样指示报文中只携带一个采样功能项时，采样功能项总数量为1。
在一具体示例中，表2中示出了一种采样指示报文包含的字段信息。
表2
Figure PCTCN2021082475-appb-000002
S130、通过NPU根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至所述CPU。
NPU，是网络设备的核心器件，具有一定的编程能力，能对网络报文进行高速处理，在本申请中，NPU接收协处理器发送的报文之后，会对多种报文进行统计分析。
NPU内部有多个统计寄存器、告警寄存器、状态寄存器，对应不同的统计功能项、告警项、状态项，指示当前的网络和设备的统计、告警和状态。
另外，NPU能访问读写一片内存空间，可以称之为采样内存空间，采样内存空间设置为存储采样值。在一具体示例中，表3中示出了采样内存空间的一种采样值存储表，其中，平均值可以是最大最小平均值。下述均以平均值是最大最小平均值为例进行解释说明，也即下文中提到的平均值均可以是最大最小平均值。
表3
Figure PCTCN2021082475-appb-000003
Figure PCTCN2021082475-appb-000004
在一示例中，通过NPU根据所述采样指示报文生成所述目标数据的采样报告报文，可以为：
通过NPU根据所述采样指示报文，更新本地存储的采样值；当所述采样指示报文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
NPU根据接收到的采样指示报文中的字段，将对应寄存器中的采样值按照序列搬移到采样内存空间中，可以是添加至如表3所示的采样值存储表中，采样值存储表中的采样值用于填充采样报告报文。
为了便于计算周期内数据平均值，NPU根据接收到的采样指示报文中的字段，将对应寄存器中的采样值按照序列搬移到采样内存空间中之后，可以对与采样编号匹配的周期内最大值和周期内最小值进行更新。进而，在需要上报平均值时，直接根据如表3所示的采样值存储表中与采样编号匹配的周期内最大值和周期内最小值进行计算即可。平均值为周期内最大值和周期内最小值的平均值。
同时，NPU通过比较采样指示报文中的报告计数值与报告间隔，判断是否已到采样报告报文发送周期。
如果采样指示报文中的采样值报告计数值与采样值报告间隔不相等，则说明未到关于采样值的采样报告报文发送周期；如果采样指示报文中的采样值报告计数值与采样值报告间隔相等，则说明已到关于采样值的采样报告报文发送周期，进而NPU生成对应的采样报告报文发送至CPU。
如果采样指示报文中的平均值报告间隔为零，则说明不需要进行平均值上报；如果采样指示报文中的平均值报告间隔不为零，则说明需要进行平均值上报，如果采样指示报文中的平均值报告计数值与平均值报告间隔不相等，则说明未到关于平均值的采样报告报文发送周期；如果采样指示报文中的平均值报告计数值与平均值报告间隔相等，则说明已到关于平均值的采样报告报文发送周期，进而NPU生成对应的采样报告报文发送至CPU。
在另一示例中，通过NPU根据所述采样指示报文生成所述目标数据的采样报告报文，可以为：
当所述采样指示报文中所述抑制冗余标识为有效值时，通过NPU根据所述采样指示报文，更新本地存储的采样值，并判断当前时刻采样值是否发生变化；通过NPU如果确定所述当前时刻采样值发生变化，则根据本地存储的采样值生成所述目标数据的采样报告报文；通过NPU如果确定所述当前时刻采样值未发生变化，则当所述采样指示报文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
当所述采样指示报文中所述抑制冗余标识为有效值时，NPU根据所述采样指示报文，更新本地存储的采样值，并判断当前时刻采样值是否发生变化：如果所述当前时刻采样值发生变化，则根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；如果所述当前时刻采样值未发生变化，则当所述采样指示报文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU。
在判断当前时刻采样值是否发生变化时，对于统计值而言，如果当前时刻采样值不为零即认为当前时刻采样值发生了变化，对于告警值和状态值而言，如果当前时刻的采样值与前一时刻的采样值不一致即认为当前时刻采样值发生了变化。进而，在抑制冗余标识为有效值时，如果当前时刻采样值发生了变化，NPU则立即向CPU发送采样报告报文，如果当前时刻采样值未发生变化，NPU则到报告周期时才向CPU发送采样报告报文。
在一具体示例中，表4中示出了一种关于采样值的采样报告报文包含的字段信息。
表4
Figure PCTCN2021082475-appb-000005
在一具体示例中，表5中示出了一种关于平均值的采样报告报文包含的字段信息。
表5
Figure PCTCN2021082475-appb-000006
CPU对接收的NPU发送的采样报告报文进行分析，获取到采样报告报文中携带的采样值，对采样值进行设定处理后即可呈现到人机界面。
本实施方式提供的技术方案，与传统的CPU主动采样的方式相比，能达到的采样频率更高(可以最快以每秒快于100次的频率对统计数据进行采样)，不会大量消耗CPU处理资源，也节省了CPU与NPU之间的消息通道带宽。进而，提高了整个网络数据的统计监控性能，为电信业务的可靠性提供了保障。
在一具体的示例中，以采样值报告为例，不需要最大最小平均值报告，也不需要冗余抑制功能。将用户参数预设到CPU，平均值报告间隔设为0，抑制冗余标识置为非真值。
CPU计算采样值报告间隔，采样值报告间隔＝采样频率/采样值报告频率，CPU计算获取一个未被占用的采样编号，并将采样参数配置到协处理器，其中，采样参数包括采样编号，采样频率，采样功能项，采样值报告间隔，平均值报告间隔，抑制冗余标识等。
协处理器根据CPU配置的采样参数，对采样频率不为0的采样功能项，以配置的采样频率生成一个携带采样编号、采样功能项、采样值报告计数值、采样值报告间隔、平均值报告计数值、平均值报告间隔、抑制冗余标识的采样指示报文，发送给NPU。
采样值报告计数值可以从1开始，每发送一个采样指示报文，如果采样功能项对应的采样值报告计数值不等于对应的采样值报告间隔，下一个发送的采样指示报文中的该采样功能项对应的采样值报告计数值递增1；如果采样功能项对应的采样值报告计数值等于对应的采样值报告间隔，下一个发送的采样指示报文中的该采样功能项对应的采样值报告计数值设置为1。
采样功能项对应的平均值报告间隔为0，采样指示报文中的平均值报告计数值可以是空值，也可以是其他值，对此不做限定。
如果有多个采样功能项是关联的，它们具有相同的采样频率和报告频率，可以同时配置，可以在采样指示报文中同时携带多个采样功能项，并且在采样指示报文中携带采样功能项总数量字段。
NPU收到采样指示报文后，识别采样指示报文中的采样功能项，读取对应的统计、告警、状态值。
NPU判断到抑制冗余标识为非真值，则按照采样指示报文将采样值搬移到匹配的采样值序列中，并更新周期内最大值和周期内最小值；NPU根据采样指示报文中的平均值报告间隔为零判断出不需要报告最大最小平均值；NPU判断采样指示报文中，采样值报告计数值和采样值报告间隔是否相等，如果采样值报告计数值和采样值报告间隔不相等，说明未到采样值报告报文发送周期，结束处理，如果采样值报告计数值和采样值报告间隔相等，说明已到采样值报告报文发送周期，生成一个采样值报告报文，填充采样值报告报文中的字段，包括采样编号、采样值等信息，发送给CPU，结束处理。
在另一具体的示例中，以采样值报告为例，还需要最大最小平均值报告，但不需要冗余抑制功能。将用户参数预设到CPU，抑制冗余标识置为非真值。
在上述示例的基础上，CPU还计算平均值报告间隔，平均值报告间隔＝采样频率/平均值报告频率。
CPU计算获取一个未被占用的采样编号，并将采样参数配置到协处理器，其中，采样参数包括采样编号，采样频率，采样功能项，采样值报告间隔，平均值报告间隔，抑制冗余标识等。
协处理器根据CPU配置的采样参数，对采样频率不为0的采样功能项，以配置的采样频率生成一个携带采样编号、采样功能项、采样值报告计数值、采样值报告间隔、平均值报告计数值、平均值报告间隔、抑制冗余标识的采样指示报文，发送给NPU。
采样值报告计数值可以从1开始，每发送一个采样指示报文，如果采样功能项对应的采样值报告计数值不等于对应的采样值报告间隔，下一个发送的采样指示报文中的该采样功能项对应的采样值报告计数值递增1；如果采样功能项对应的采样值报告计数值等于对应的采样值报告间隔，下一个发送的采样指示报文中的该采样功能项对应的采样值报告计数值设置为1。
如果采样功能项对应的平均值报告间隔不为0，则采样指示报文中的平均值报告计数值从1开始，每发送一个采样指示报文，如果采样功能项对应的平均值报告计数值不等于对应的平均值报告间隔，下一个发送的采样指示报文中的该采样功能项对应的平均值报告计数值递增1；如果采样功能项对应的平均值报告计数值等于对应的平均值报告间隔，下一个发送的采样指示报文中的该采样功能项对应的平均值报告计数值设置为1。
如果有多个采样功能项是关联的，它们具有相同的采样频率和报告频率， 可以同时配置，可以在采样指示报文中同时携带多个采样功能项，并且在采样指示报文中携带采样功能项总数量字段。
NPU收到采样指示报文后，识别采样指示报文中的采样功能项，读取对应的统计、告警、状态值。
NPU判断到抑制冗余标识为非真值，则按照采样指示报文将采样值搬移到匹配的采样值序列中，并更新周期内最大值和周期内最小值；NPU根据采样指示报文中的平均值报告间隔不为零判断出需要报告最大最小平均值，NPU判断采样指示报文中，平均值报告计数值和平均值报告间隔是否相等，如果平均值报告计数值和平均值报告间隔不相等，说明未到平均值报告报文发送周期，如果平均值报告计数值和平均值报告间隔相等，说明已到平均值报告报文发送周期，生成一个平均值报告报文，填充平均值报告报文中的字段，包括采样编号、平均值等信息，发送给CPU，并清空存储空间中对应的周期内最大值和周期内最小值；NPU判断采样指示报文中，采样值报告计数值和采样值报告间隔是否相等，如果采样值报告计数值和采样值报告间隔不相等，说明未到采样值报告报文发送周期，结束处理，如采样值报告计数值和采样值报告间隔果相等，说明已到采样值报告报文发送周期，生成一个采样值报告报文，填充采样值报告报文中的字段，包括采样编号、采样值等信息，发送给CPU，结束处理。
在又一具体的示例中，以采样值报告为例，还需要冗余抑制功能，将用户参数预设到CPU，抑制冗余标识置为真值。
NPU收到采样指示报文后，识别采样指示报文中的采样功能项，读取对应的统计、告警、状态值。
NPU判断到抑制冗余标识为真值，则按照采样指示报文将采样值搬移到匹配的采样值序列中，并更新周期内最大值和周期内最小值，判断当前时刻采样值是否发生变化，对于统计值而言，如果当前时刻采样值不为零即认为当前时刻采样值发生了变化，对于告警值和状态值而言，如果当前时刻的采样值与前一时刻的采样值不一致即认为当前时刻采样值发生了变化。
如果当前时刻采样值发生了变化，则生成一个采样值报告报文，填充采样值报告报文中的字段，包括采样编号、采样值等信息，发送给CPU，结束处理；如果当前时刻采样值未发生变化，则判断采样指示报文中，采样值报告计数值和采样值报告间隔是否相等，如果采样值报告计数值和采样值报告间隔不相等，说明未到采样值报告报文发送周期，结束处理，如果采样值报告计数值和采样值报告间隔相等，说明已到采样值报告报文发送周期，生成一个采样值报告报文，填充采样值报告报文中的字段，包括采样编号、采样值等信息，发送给CPU，结束处理。
图4为本申请实施例提供的另一种网络设备的数据采样方法的流程图。
S210、CPU计算配置采样参数，包括采样编号、采样频率、采样值报告间隔、平均值报告间隔、抑制冗余标识。
S220、CPU将采样参数配置到协处理器的采样参数配置表。
S230、协处理器根据采样参数配置表生成采样指示报文，发送至NPU。
S240、NPU根据采样指示报文读取对应的采样数值，并处理。
S250、NPU根据采样指示报文判断抑制冗余标识是否为真值，若抑制冗余标识为真值，则执行S2110，若抑制冗余标识不为真值，则执行S260。
S260、NPU将采样值按照采样编号搬移到存储空间，更新周期内最大值和周期内最小值。
S270、NPU判断是否平均值报告间隔非零且平均值报告计数值与平均值报告间隔相等，若平均值报告间隔非零且平均值报告计数值与平均值报告间隔相等，则执行S2100，若平均值报告间隔为零或平均值报告计数值不与平均值报告间隔相等，则执行S280。
S280、NPU判断是否采样值报告计数值与采样值报告间隔相等，若采样值报告计数值与采样值报告间隔相等，则执行S290，若采样值报告计数值不与采样值报告间隔相等，则执行S2130。
S290、NPU生成采样值报告报文，填充采样值，发送至CPU。
S2100、NPU生成平均值报告报文，填充最大最小平均值，发送至CPU，清除存储空间中的周期内最大值和周期内最小值。
S2110、NPU将当前时刻采样值搬移到对应的存储空间。
S2120、NPU判断当前时刻采样值是否发生了变化，若当前时刻采样值发生了变化，则执行S290，若当前时刻采样值未发生了变化，则执行S280。
S2130、结束处理。
本实施方式提供的技术方案，通过协处理器快速发送采样消息到NPU中进行快速采样，NPU将采样值搬移到存储空间中，再定期汇总向CPU发送报告消息的方式来实现快速对网络设备的性能、告警以及状态的采样。与传统的CPU主动采样的方式相比，上述技术方案能达到的采样频率更高(可以最快以每秒快于100次的频率对统计数据进行采样)，不会大量消耗CPU处理资源，也节省了CPU与NPU之间的消息通道带宽。进而，提高了整个网络数据的统计监控性能，为电信业务的可靠性提供了保障。
在一个示例性实施方式中，图5为本申请实施例提供的另一种网络设备的数据采样方法的流程示意图。该方法适用于以高频率对网络设备处理的业务(或性能)统计、实时告警、实时状态等进行采样的情况。该方法可以由本申请提供的应用于CPU中的网络设备的数据采样装置执行，该网络设备的数据采样装置可以由软件和/或硬件实现，并集成在CPU中。
如图5所示，本申请提供的应用于CPU中的网络设备的数据采样方法，包括：
S310、将目标数据的采样参数配置至协处理器，其中，所述采样参数用于所述协处理器以设定频率生成采样指示报文并将所述采样指示报文发送至NPU，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至CPU。
在一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔。
所述报告间隔包括：采样值报告间隔和/或平均值报告间隔。其中，采样值报告间隔，定义了每采样多少次就上报一次数据采样值；平均值报告间隔，定义了每采样多少次就上报一次数据平均值。
采样值报告间隔根据采样值报告频率和所述采样频率确定，其中，采样值报告频率为用户设置的参数，定义了每秒上报数据采样值的次数，例如，采样值报告间隔＝采样频率/采样值报告频率。
平均值报告间隔根据平均值报告频率和所述采样频率确定，其中，平均值报告频率为用户设置的参数，定义了每秒上报数据平均值的次数，例如，平均值报告间隔＝采样频率/平均值报告频率。
在一示例中，平均值可以指关于所有采样值的平均值。在另一示例中，平均值可以指关于最大采样值和最小采样值的平均值，即最大最小平均值。本实施例中以平均值为最大最小平均值为例进行解释说明。
在另一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔、抑制冗余标识。
抑制冗余标识，为用户设置的参数，用于标识采样过程中是否需要冗余抑制功能。抑制冗余标识为有效值，对应于采样过程中需要冗余抑制功能。
协处理器读取CPU配置的目标数据的采样参数，根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至NPU。
在一示例中，协处理器根据所述采样参数以设定频率生成采样指示报文可以为：协处理器根据所述采样参数，对采样频率不为零的采样功能项，以所述采样频率生成采样指示报文。
在一示例中，协处理器内部可以维护一张逻辑表格，此处可以称之为采样参数配置表，用于表述当前需要进行的采样功能项以及对应的参数。快速采样参数配置表，是协处理器能够访问的逻辑表格，可以由CPU进行配置，也即CPU可以将采样参数配置到该采样参数配置表中，从而控制协处理器基于采样参数配置表中的采样参数生成采样指示报文并发送。
在一示例中，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔；或者，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔、抑制冗余标识；当所述报告间隔包括采样值报告间隔时，所述报告计数值包括采样值报告计数值；在依次发送的所述采样指示报文中，所述采样值报告计数值由初始值依次递增，增至所述采样值报告间隔时重置为所述初始值；当所述报告间隔包括平均值报告间隔时，所述报告计数值包括平均值报告计数值；对所述平均值报告间隔不为零的采样功能项，在依次发送的所述采样指示报文中，所述平均值报告计数值由初始值依次递增，增至所述平均值报告间隔时重置为所述初始值。
在上述示例的基础上，所述采样指示报文中还可以包括：采样功能项总数量，所述采样功能项总数量用于指示所述采样指示报文中包括的采样编号以及采样功能项的组数量，不同组中的采样功能项的采样频率和报告间隔均相同。
如果多个采样功能项是关联的，具有相同的采样频率和报告频率，则可以同时进行配置，进而，协处理器可以在一个采样指示报文中携带多个采样功能项，并在该采样指示报文中携带采样功能项总数量。
在本申请中，NPU接收网络上的报文之后，会对多种报文进行统计分析。
NPU内部有多个统计寄存器、告警寄存器、状态寄存器，对应不同的统计功能项、告警项、状态项，指示当前的网络和设备的统计、告警和状态。
另外，NPU能访问读写一片内存空间，称之为采样内存空间，采样内存空间设置为存储采样值，例如通过采样值存储表的形式存储采样值。
NPU根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至所述CPU。
在一示例中，NPU根据所述采样指示报文生成所述目标数据的采样报告报文，可以为：
NPU根据所述采样指示报文，更新本地存储的采样值；当所述采样指示报 文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
在另一示例中，NPU根据所述采样指示报文生成所述目标数据的采样报告报文，可以为：
当所述采样指示报文中所述抑制冗余标识为有效值时，NPU根据所述采样指示报文，更新本地存储的采样值，并判断当前时刻采样值是否发生变化；如果所述当前时刻采样值发生变化，则根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；如果所述当前时刻采样值未发生变化，则当所述采样指示报文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
S320、接收所述NPU发送的所述目标数据的采样报告报文。
CPU接收所述NPU发送的所述目标数据的采样报告报文，并分析，获取到采样报告报文中携带的采样值，对采样值进行设定处理后即可呈现到人机界面。
本实施方式未尽解释之处请参见前述实施方式，在此不再赘述。
本实施方式提供的技术方案，与传统的CPU主动采样的方式相比，能达到的采样频率更高(可以最快以每秒快于100次的频率对统计数据进行采样)，不会大量消耗CPU处理资源，也节省了CPU与NPU之间的消息通道带宽。进而，提高了整个网络数据的统计监控性能，为电信业务的可靠性提供了保障。
在一个示例性实施方式中，图6为本申请实施例提供的另一种网络设备的数据采样方法的流程示意图。该方法适用于以高频率对网络设备处理的业务(或性能)统计、实时告警、实时状态等进行采样的情况。该方法可以由本申请提供的应用于协处理器中的网络设备的数据采样装置执行，该网络设备的数据采样装置可以由软件和/或硬件实现，并集成在协处理器中。
可选的，协处理器为外挂的FPGA，或者为NPU中的包发生器，本实施例对此不做限定，网络设备中具有简单的计算处理能力和快速发包能力的装置均可以作为协处理器。
如图6所示，本申请提供的应用于协处理器中的网络设备的数据采样方法，包括：
S410、读取CPU配置的目标数据的采样参数。
在一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔。
所述报告间隔包括：采样值报告间隔和/或平均值报告间隔。其中，采样值报告间隔，定义了每采样多少次就上报一次数据采样值；平均值报告间隔，定义了每采样多少次就上报一次数据平均值。
采样值报告间隔根据采样值报告频率和所述采样频率确定，其中，采样值报告频率为用户设置的参数，定义了每秒上报数据采样值的次数，例如，采样值报告间隔＝采样频率/采样值报告频率。
平均值报告间隔根据平均值报告频率和所述采样频率确定，其中，平均值报告频率为用户设置的参数，定义了每秒上报数据平均值的次数，例如，平均值报告间隔＝采样频率/平均值报告频率。
在一示例中，平均值可以指关于所有采样值的平均值。在另一示例中，平均值可以指关于最大采样值和最小采样值的平均值，即最大最小平均值。本实施例中以平均值为最大最小平均值为例进行解释说明。
在另一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔、抑制冗余标识。
抑制冗余标识，为用户设置的参数，用于标识采样过程中是否需要冗余抑制功能。抑制冗余标识为有效值，对应于采样过程中需要冗余抑制功能。
S420、根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至NPU，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至CPU。
在一示例中，协处理器根据所述采样参数以设定频率生成采样指示报文，可以为：协处理器根据所述采样参数，对采样频率不为零的采样功能项，以所述采样频率生成采样指示报文。
在一示例中，协处理器内部可以维护一张逻辑表格，此处可以称之为采样参数配置表，用于表述当前需要进行的采样功能项以及对应的参数。快速采样参数配置表，是协处理器能够访问的逻辑表格，可以由CPU进行配置，也即CPU可以将采样参数配置到该采样参数配置表中，从而控制协处理器基于采样参数配置表中的采样参数进行采样指示报文的发送。
在一示例中，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔；或者，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔、抑制冗余标识；当所述报告间隔包括采样值报告间隔时，所述报告计数值包括采样值报告计数值；在依次发送的所述 采样指示报文中，所述采样值报告计数值由初始值依次递增，增至所述采样值报告间隔时重置为所述初始值；当所述报告间隔包括平均值报告间隔时，所述报告计数值包括平均值报告计数值；对所述平均值报告间隔不为零的采样功能项，在依次发送的所述采样指示报文中，所述平均值报告计数值由初始值依次递增，增至所述平均值报告间隔时重置为所述初始值。
在上述示例的基础上，所述采样指示报文中还可以包括：采样功能项总数量，所述采样功能项总数量用于指示所述采样指示报文中包括的采样编号以及采样功能项的组数量，不同组中的采样功能项的采样频率和报告间隔均相同。
如果多个采样功能项是关联的，具有相同的采样频率和报告频率，则可以同时进行配置，进而，协处理器可以在一个采样指示报文中携带多个采样功能项，并在该采样指示报文中携带采样功能项总数量。
在本申请中，NPU接收网络上的报文之后，会对多种报文进行统计分析。
NPU内部有多个统计寄存器、告警寄存器、状态寄存器，对应不同的统计功能项、告警项、状态项，指示当前的网络和设备的统计、告警和状态。
另外，NPU能访问读写一片内存空间，称之为采样内存空间，采样内存空间设置为存储采样值，例如通过采样值存储表的形式存储采样值。
NPU根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至所述CPU。
在一示例中，NPU根据所述采样指示报文生成所述目标数据的采样报告报文，可以为：
NPU根据所述采样指示报文，更新本地存储的采样值；当所述采样指示报文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
在另一示例中，NPU根据所述采样指示报文生成所述目标数据的采样报告报文，可以为：
当所述采样指示报文中所述抑制冗余标识为有效值时，NPU根据所述采样指示报文，更新本地存储的采样值，并判断当前时刻采样值是否发生变化；如果所述当前时刻采样值发生变化，则根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；如果所述当前时刻采样值未发生变化，则当所述采样指示报文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
CPU接收所述NPU发送的所述目标数据的采样报告报文，并分析，获取到采样报告报文中携带的采样值，对采样值进行设定处理后即可呈现到人机界面。
本实施方式未尽解释之处请参见前述实施方式，在此不再赘述。
本实施方式提供的技术方案，与传统的CPU主动采样的方式相比，能达到的采样频率更高(可以最快以每秒快于100次的频率对统计数据进行采样)，不会大量消耗CPU处理资源，也节省了CPU与NPU之间的消息通道带宽。进而，提高了整个网络数据的统计监控性能，为电信业务的可靠性提供了保障。
在一个示例性实施方式中，图7为本申请实施例提供的另一种网络设备的数据采样方法的流程示意图。该方法适用于以高频率对网络设备处理的业务(或性能)统计、实时告警、实时状态等进行采样的情况。该方法可以由本申请提供的应用于NPU中的网络设备的数据采样装置执行，该网络设备的数据采样装置可以由软件和/或硬件实现，并集成在NPU中。
如图7所示，本申请提供的应用于NPU中的网络设备的数据采样方法，包括：
S510、接收协处理器发送的采样指示报文，所述采样指示报文为所述协处理器根据CPU发送的目标数据的采样参数以设定频率生成并发送的。
CPU将目标数据的采样参数配置至协处理器，在一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔。
所述报告间隔包括：采样值报告间隔和/或平均值报告间隔。其中，采样值报告间隔，定义了每采样多少次就上报一次数据采样值；平均值报告间隔，定义了每采样多少次就上报一次数据平均值。
采样值报告间隔根据采样值报告频率和所述采样频率确定，其中，采样值报告频率为用户设置的参数，定义了每秒上报数据采样值的次数，例如，采样值报告间隔＝采样频率/采样值报告频率。
平均值报告间隔根据平均值报告频率和所述采样频率确定，其中，平均值报告频率为用户设置的参数，定义了每秒上报数据平均值的次数，例如，平均值报告间隔＝采样频率/平均值报告频率。
在一示例中，平均值可以指关于所有采样值的平均值。在另一示例中，平均值可以指关于最大采样值和最小采样值的平均值，即最大最小平均值。本实施例中以平均值为最大最小平均值为例进行解释说明。
在另一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频 率、报告间隔、抑制冗余标识。
抑制冗余标识，为用户设置的参数，用于标识采样过程中是否需要冗余抑制功能。抑制冗余标识为有效值，对应于采样过程中需要冗余抑制功能。
协处理器根据所述采样参数以设定频率生成采样指示报文发送至NPU。
在一示例中，协处理器根据所述采样参数以设定频率生成采样指示报文，可以为：协处理器根据所述采样参数，对采样频率不为零的采样功能项，以所述采样频率生成采样指示报文。
协处理器内部维护一张逻辑表格，可以称之为采样参数配置表，用于表述当前需要进行的采样功能项以及对应的参数。快速采样参数配置表，是协处理器能够访问的逻辑表格，可以由CPU进行配置，也即CPU可以将采样参数配置到该采样参数配置表中，从而控制协处理器基于采样参数配置表中的采样参数进行采样指示报文的发送。
在一示例中，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔；或者，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔、抑制冗余标识；当所述报告间隔包括采样值报告间隔时，所述报告计数值包括采样值报告计数值；在依次发送的所述采样指示报文中，所述采样值报告计数值由初始值依次递增，增至所述采样值报告间隔时重置为所述初始值；当所述报告间隔包括平均值报告间隔时，所述报告计数值包括平均值报告计数值；对所述平均值报告间隔不为零的采样功能项，在依次发送的所述采样指示报文中，所述平均值报告计数值由初始值依次递增，增至所述平均值报告间隔时重置为所述初始值。
在上述示例的基础上，所述采样指示报文中还可以包括：采样功能项总数量，所述采样功能项总数量用于指示所述采样指示报文中包括的采样编号以及采样功能项的组数量，不同组中的采样功能项的采样频率和报告间隔均相同。
如果多个采样功能项是关联的，具有相同的采样频率和报告频率，则可以同时进行配置，进而，协处理器可以在一个采样指示报文中携带多个采样功能项，并在该采样指示报文中携带采样功能项总数量。
S520、根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至CPU。
在本申请中，NPU接收网络上的报文之后，会对多种报文进行统计分析。
NPU内部有多个统计寄存器、告警寄存器、状态寄存器，对应不同的统计功能项、告警项、状态项，指示当前的网络和设备的统计、告警和状态。
另外，NPU能访问读写一片内存空间，称之为采样内存空间，采样内存空间设置为存储采样值，例如通过采样值存储表的形式存储采样值。
在一示例中，NPU根据所述采样指示报文生成所述目标数据的采样报告报文，可以为：
根据所述采样指示报文，更新本地存储的采样值；当所述采样指示报文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
在另一示例中，NPU根据所述采样指示报文生成所述目标数据的采样报告报文，可以为：
当所述采样指示报文中所述抑制冗余标识为有效值时，NPU根据所述采样指示报文，更新本地存储的采样值，并判断当前时刻采样值是否发生变化；如果所述当前时刻采样值发生变化，则根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；如果所述当前时刻采样值未发生变化，则当所述采样指示报文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
CPU接收所述NPU发送的所述目标数据的采样报告报文，并分析，获取到采样报告报文中携带的采样值，对采样值进行设定处理后即可呈现到人机界面。
本实施方式未尽解释之处请参见前述实施方式，在此不再赘述。
本实施方式提供的技术方案，与传统的CPU主动采样的方式相比，能达到的采样频率更高(可以最快以每秒快于100次的频率对统计数据进行采样)，不会大量消耗CPU处理资源，也节省了CPU与NPU之间的消息通道带宽。进而，提高了整个网络数据的统计监控性能，为电信业务的可靠性提供了保障。
本实施例还提供了一种网络设备的数据采样装置，图8为本申请实施例提供的一种网络设备的数据采样装置的结构示意图。如图8所示，本申请实施例提供的应用于CPU中的网络设备的数据采样装置，可以由软件和/或硬件实现，并集成在CPU中，该装置包括：采样参数配置模块610和采样报告报文接收模块620。
采样参数配置模块610，设置为将目标数据的采样参数配置至协处理器，其中，所述采样参数用于所述协处理器以设定频率生成采样指示报文并将所述采样指示报文发送至NPU，所述采样指示报文用于所述NPU生成所述目标数据的 采样报告报文并将所述采样报告报文发送至CPU；采样报告报文接收模块620，设置为接收所述NPU发送的所述目标数据的采样报告报文。
本实施例提供的技术方案，与传统的CPU主动采样的方式相比，能达到的采样频率更高(可以最快以每秒快于100次的频率对统计数据进行采样)，不会大量消耗CPU处理资源，也节省了CPU与NPU之间的消息通道带宽。进而，提高了整个网络数据的统计监控性能，为电信业务的可靠性提供了保障。
在一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔；或者，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔、抑制冗余标识；其中，所述报告间隔包括：采样值报告间隔和/或平均值报告间隔，所述采样值报告间隔根据采样值报告频率和所述采样频率确定，所述平均值报告间隔根据平均值报告频率和所述采样频率确定。
在一示例中，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔；或者，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔、抑制冗余标识；当所述报告间隔包括采样值报告间隔时，所述报告计数值包括采样值报告计数值；在依次发送的所述采样指示报文中，所述采样值报告计数值由初始值依次递增，增至所述采样值报告间隔时重置为所述初始值；当所述报告间隔包括平均值报告间隔时，所述报告计数值包括平均值报告计数值；对所述平均值报告间隔不为零的采样功能项，在依次发送的所述采样指示报文中，所述平均值报告计数值由初始值依次递增，增至所述平均值报告间隔时重置为所述初始值。
在一示例中，所述采样指示报文中还包括：采样功能项总数量，所述采样功能项总数量用于指示所述采样指示报文中包括的采样编号以及采样功能项的组数量，不同组中的采样功能项的采样频率和报告间隔均相同。
本实施例提供的应用于CPU中的网络设备的数据采样装置设置为实现如本申请实施例所述的应用于CPU中的网络设备的数据采样方法，本实施例提供的应用于CPU中的网络设备的数据采样装置实现原理和技术效果与本申请实施例所述的应用于CPU中的网络设备的数据采样方法类似，此处不再赘述。
本实施例还提供了一种网络设备的数据采样装置，图9为本申请实施例提供的另一种网络设备的数据采样装置的结构示意图。如图9所示，本申请实施例提供的应用于协处理器中的网络设备的数据采样装置，可以由软件和/或硬件实现，并集成在协处理器中，该装置包括：采样参数读取模块710和采样指示报文发送模块720。
采样参数读取模块710，设置为读取CPU配置的目标数据的采样参数；采样指示报文发送模块720，设置为根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至NPU，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至CPU。
本实施例提供的技术方案，与传统的CPU主动采样的方式相比，能达到的采样频率更高(可以最快以每秒快于100次的频率对统计数据进行采样)，不会大量消耗CPU处理资源，也节省了CPU与NPU之间的消息通道带宽。进而，提高了整个网络数据的统计监控性能，为电信业务的可靠性提供了保障。
在一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔；或者，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔、抑制冗余标识；其中，所述报告间隔包括：采样值报告间隔和/或平均值报告间隔，所述采样值报告间隔根据采样值报告频率和所述采样频率确定，所述平均值报告间隔根据平均值报告频率和所述采样频率确定。
在一示例中，采样指示报文发送模块720，设置为根据所述采样参数，对采样频率不为零的采样功能项，以所述采样频率生成采样指示报文，并将所述采样指示报文发送至NPU；所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔；或者，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔、抑制冗余标识；当所述报告间隔包括采样值报告间隔时，所述报告计数值包括采样值报告计数值；在依次发送的所述采样指示报文中，所述采样值报告计数值由初始值依次递增，增至所述采样值报告间隔时重置为所述初始值；当所述报告间隔包括平均值报告间隔时，所述报告计数值包括平均值报告计数值；对所述平均值报告间隔不为零的采样功能项，在依次发送的所述采样指示报文中，所述平均值报告计数值由初始值依次递增，增至所述平均值报告间隔时重置为所述初始值。
在一示例中，所述采样指示报文中还包括：采样功能项总数量，所述采样功能项总数量用于指示所述采样指示报文中包括的采样编号以及采样功能项的组数量，不同组中的采样功能项的采样频率和报告间隔均相同。
本实施例提供的应用于协处理器中的网络设备的数据采样装置设置为实现如本申请实施例所述的应用于协处理器中的网络设备的数据采样方法，本实施例提供的应用于协处理器中的网络设备的数据采样装置实现原理和技术效果与本申请实施例所述的应用于协处理器中的网络设备的数据采样方法类似，此处不再赘述。
本实施例还提供了一种网络设备的数据采样装置，图10为本申请实施例提供的另一种网络设备的数据采样装置的结构示意图。如图10所示，本申请实施例提供的应用于NPU中的网络设备的数据采样装置，可以由软件和/或硬件实现，并集成在NPU中，该装置包括：采样指示报文接收模块810和采样报告报文发送模块820。
采样指示报文接收模块810，设置为接收协处理器发送的采样指示报文，所述采样指示报文为所述协处理器根据CPU发送的目标数据的采样参数以设定频率生成并发送的；采样报告报文发送模块820，设置为根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至CPU。
本实施例提供的技术方案，与传统的CPU主动采样的方式相比，能达到的采样频率更高(可以最快以每秒快于100次的频率对统计数据进行采样)，不会大量消耗CPU处理资源，也节省了CPU与NPU之间的消息通道带宽。进而，提高了整个网络数据的统计监控性能，为电信业务的可靠性提供了保障。
在一示例中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔；或者，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔、抑制冗余标识；其中，所述报告间隔包括：采样值报告间隔和/或平均值报告间隔，所述采样值报告间隔根据采样值报告频率和所述采样频率确定，所述平均值报告间隔根据平均值报告频率和所述采样频率确定。
在一示例中，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔；或者，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔、抑制冗余标识；当所述报告间隔包括采样值报告间隔时，所述报告计数值包括采样值报告计数值；在依次发送的所述采样指示报文中，所述采样值报告计数值由初始值依次递增，增至所述采样值报告间隔时重置为所述初始值；当所述报告间隔包括平均值报告间隔时，所述报告计数值包括平均值报告计数值；对所述平均值报告间隔不为零的采样功能项，在依次发送的所述采样指示报文中，所述平均值报告计数值由初始值依次递增，增至所述平均值报告间隔时重置为所述初始值。
在一示例中，采样报告报文发送模块820，设置为根据所述采样指示报文，更新本地存储的采样值；当所述采样指示报文中的报告计数值等于报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
在一示例中，采样报告报文发送模块820，设置为当所述采样指示报文中的抑制冗余标识为有效值时，根据所述采样指示报文，更新本地存储的采样值，并判断当前时刻采样值是否发生变化；如果所述当前时刻采样值发生变化，则 根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；如果所述当前时刻采样值未发生变化，则当所述采样指示报文中的所述报告计数值等于所述报告间隔时，根据本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；其中，所述采样报告报文包括采样值报告报文和/或平均值报告报文。
本实施例提供的应用于NPU中的网络设备的数据采样装置设置为实现如本申请实施例所述的应用于NPU中的网络设备的数据采样方法，本实施例提供的应用于NPU中的网络设备的数据采样装置实现原理和技术效果与本申请实施例所述的应用于NPU中的网络设备的数据采样方法类似，此处不再赘述。
本申请实施例提供了一种CPU，图11为本申请实施例提供的一种CPU的结构示意图。如图11所示，本申请提供的CPU，包括：一个或多个处理单元910和存储单元920；该CPU的处理单元910可以是一个或多个，图11中以一个处理单元910为例；存储单元920设置为存储一个或多个程序；所述一个或多个程序被所述一个或多个处理单元910执行，使得所述一个或多个处理单元910实现如本申请实施例中所述的应用于CPU中的网络设备的数据采样方法。
CPU中的处理单元910、存储单元920可以通过总线或其他方式连接，图11中以通过总线连接为例。
存储单元920作为一种计算机可读存储介质，可设置为存储软件程序、计算机可执行程序以及模块，如本申请实施例所述应用于CPU中的网络设备的数据采样方法对应的程序指令/模块(例如，附图8所示的应用于CPU中的网络设备的数据采样装置中的采样参数配置模块610和采样报告报文接收模块620)。存储单元920可包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序；存储数据区可存储根据设备的使用所创建的数据等。此外，存储单元920可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。
本申请实施例提供了一种协处理器，图12为本申请实施例提供的一种协处理器的结构示意图。如图12所示，本申请提供的协处理器，包括：一个或多个处理单元1010和存储单元1020；该协处理器的处理单元1010可以是一个或多个，图12中以一个处理单元1010为例；存储单元1020设置为存储一个或多个程序；所述一个或多个程序被所述一个或多个处理单元1010执行，使得所述一 个或多个处理单元1010实现如本申请实施例中所述的应用于协处理器中的网络设备的数据采样方法。
协处理器中的处理单元1010、存储单元1020可以通过总线或其他方式连接，图12中以通过总线连接为例。
存储单元1020作为一种计算机可读存储介质，可设置为存储软件程序、计算机可执行程序以及模块，如本申请实施例所述应用于协处理器中的网络设备的数据采样方法对应的程序指令/模块(例如，附图9所示的应用于协处理器中的网络设备的数据采样装置中的采样参数读取模块710和采样指示报文发送模块720)。存储单元1020可包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序；存储数据区可存储根据设备的使用所创建的数据等。此外，存储单元1020可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。
本申请实施例提供了一种NPU，图13为本申请实施例提供的一种协处理器的结构示意图。如图13所示，本申请提供的NPU，包括：一个或多个处理单元1110和存储单元1120；该NPU的处理单元1110可以是一个或多个，图13中以一个处理单元1110为例；存储单元1120设置为存储一个或多个程序；所述一个或多个程序被所述一个或多个处理单元1110执行，使得所述一个或多个处理单元1110实现如本申请实施例中所述的应用于NPU中的网络设备的数据采样方法。
协处理器中的处理单元1110、存储单元1120可以通过总线或其他方式连接，图13中以通过总线连接为例。
存储单元1120作为一种计算机可读存储介质，可设置为存储软件程序、计算机可执行程序以及模块，如本申请实施例所述应用于NPU中的网络设备的数据采样方法对应的程序指令/模块(例如，附图10所示的应用于NPU中的网络设备的数据采样装置中的采样指示报文接收模块810和采样报告报文发送模块820)。存储单元1120可包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序；存储数据区可存储根据设备的使用所创建的数据等。此外，存储单元1120可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。
本申请实施例提供了一种网络设备，如图2所示，一种网络设备，包括：CPU、协处理器和NPU。
所述CPU执行如本申请任意实施例所述的应用于CPU中的网络设备的数据采样方法；所述协处理器执行如本申请任意实施例所述的应用于协处理器中的网络设备的数据采样方法；所述NPU执行如本申请任意实施例所述的应用于NPU中的网络设备的数据采样方法。
在一示例中，所述协处理器为网络设备外挂的FPGA。
在一示例中，所述协处理器为所述NPU中的包发生器。
本实施例未尽解释之处请参见前述实施例，在此不再赘述。
本申请实施例还提供一种存储介质，所述存储介质存储有计算机程序，所述计算机程序被处理器执行时实现本申请实施例中所述的应用于CPU中的网络设备的数据采样方法，或者实现本申请实施例中所述的应用于协处理器中的网络设备的数据采样方法，或者实现本申请实施例中所述的应用于NPU中的网络设备的数据采样方法，或者实现本申请实施例中所述的应用于网络设备中的网络设备的数据采样方法。
应用于CPU中的网络设备的数据采样方法，包括：
将目标数据的采样参数配置至协处理器，其中，所述采样参数用于所述协处理器以设定频率生成采样指示报文并将所述采样指示报文发送至NPU，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至CPU；接收所述NPU发送的所述目标数据的采样报告报文。
应用于协处理器中的网络设备的数据采样方法，包括：
读取CPU配置的目标数据的采样参数；根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至NPU，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至CPU。
应用于NPU中的网络设备的数据采样方法，包括：
接收协处理器发送的采样指示报文，所述采样指示报文为所述协处理器根据CPU发送的目标数据的采样参数以设定频率生成并发送的；根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至CPU。
应用于网络设备中的网络设备的数据采样方法，包括：
通过CPU将目标数据的采样参数配置至协处理器；通过所述协处理器根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至 NPU；通过NPU根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至所述CPU。
可选的，该计算机可执行指令在由计算机处理器执行时还可以用于实现本申请实施例中任一所述的应用于CPU中的网络设备的数据采样方法，或者实现本申请实施例中任一所述的应用于协处理器中的网络设备的数据采样方法，或者实现本申请实施例中任一所述的应用于NPU中的网络设备的数据采样方法，或者实现本申请实施例中任一所述的应用于网络设备中的网络设备的数据采样方法。
通过以上关于实施方式的描述，本申请可借助软件及必需的通用硬件来实现，也可以通过硬件实现。本申请的技术方案本质上可以以软件产品的形式体现出来，该计算机
软件产品可以存储在计算机可读存储介质中，如计算机的软盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、闪存(FLASH)、硬盘或光盘等，包括多个指令用以使得一台通信设备(可以是个人计算机，服务器，或者网络设备等)执行本申请实施例所述的方法。
上述应用于CPU中的网络设备的数据采样装置，或者应用于协处理器中的应用于协处理器中的网络设备的数据采样装置，或者应用于NPU中的网络设备的数据采样装置的实施例中，所包括的多个单元和模块只是按照功能逻辑进行划分的，但并不局限于上述的划分，只要能够实现相应的功能即可；另外，多个功能单元的名称也只是为了便于相互区分，并不用于限制本申请的保护范围。
一般来说，本申请的多种实施例可以在硬件或专用电路、软件、逻辑或其任何组合中实现。例如，一些方面可以被实现在硬件中，而其它方面可以被实现在可以被控制器、微处理器或其它计算装置执行的固件或软件中，尽管本申请不限于此。
本申请的实施例可以通过移动装置的数据处理器执行计算机程序指令来实现，例如在处理器实体中，或者通过硬件，或者通过软件和硬件的组合。计算机程序指令可以是汇编指令、指令集架构(Instruction Set Architecture，ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码。
本申请附图中的任何逻辑流程的框图可以表示程序步骤，或者可以表示相互连接的逻辑电路、模块和功能，或者可以表示程序步骤与逻辑电路、模块和 功能的组合。计算机程序可以存储在存储器上。存储器可以具有任何适合于本地技术环境的类型并且可以使用任何适合的数据存储技术实现，例如但不限于只读存储器(ROM)、随机访问存储器(RAM)、光存储器装置和系统(数码多功能光碟(Digital Video Disc，DVD)或光盘(Compact Disk，CD))等。计算机可读介质可以包括非瞬时性存储介质。数据处理器可以是任何适合于本地技术环境的类型，例如但不限于通用计算机、专用计算机、微处理器、数字信号处理器(Digital Signal Processing，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、FPGA以及基于多核处理器架构的处理器。

Claims (23)

  1. 一种网络设备的数据采样方法，应用于网络设备，包括：
    通过中央处理单元CPU将目标数据的采样参数配置至协处理器；
    通过所述协处理器根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至网络处理单元NPU；
    通过所述NPU根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至所述CPU。
  2. 根据权利要求1所述的方法，其中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔；或者，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔、抑制冗余标识；
    其中，所述报告间隔包括：采样值报告间隔和平均值报告间隔中的至少之一，所述采样值报告间隔根据采样值报告频率和所述采样频率确定，所述平均值报告间隔根据平均值报告频率和所述采样频率确定。
  3. 根据权利要求2所述的方法，其中，所述通过所述协处理器根据所述采样参数以设定频率生成采样指示报文，包括：
    通过所述协处理器根据所述采样参数，对所述采样参数中的采样频率不为零的采样功能项，以不为零的所述采样频率生成采样指示报文；
    其中，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔；或者，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔、抑制冗余标识；
    在所述报告间隔包括采样值报告间隔的情况下，所述报告计数值包括采样值报告计数值；依次发送的采样指示报文中的采样值报告计数值由初始值依次递增至所述采样值报告间隔，在发送的采样指示报文中的采样值报告计数值为所述采样值报告间隔的情况下，重置下一次发送的采样指示报文中的采样值报告计数值为所述初始值；
    在所述报告间隔包括平均值报告间隔的情况下，所述报告计数值包括平均值报告计数值；对所述平均值报告间隔不为零的采样功能项，依次发送的采样指示报文中的平均值报告计数值由初始值依次递增至所述平均值报告间隔，在发送的采样指示报文中的平均值报告计数值为所述平均值报告间隔的情况下，重置下一次发送的采样指示报文中的平均值报告计数值为所述初始值。
  4. 根据权利要求3所述的方法，其中，所述采样指示报文中还包括：采样功能项总数量，所述采样功能项总数量用于指示所述采样指示报文中包括的采样编号以及采样功能项的组数量；
    不同组中的采样功能项的采样频率和报告间隔均相同。
  5. 根据权利要求3所述的方法，其中，所述通过所述NPU根据所述采样指示报文生成所述目标数据的采样报告报文，包括：
    通过所述NPU根据所述采样指示报文，更新本地存储的采样值；
    在所述采样指示报文中的所述报告计数值等于所述报告间隔的情况下，通过所述NPU根据更新后的本地存储的采样值生成所述目标数据的采样报告报文；
    其中，所述采样报告报文包括采样值报告报文和平均值报告报文中的至少之一。
  6. 根据权利要求3所述的方法，其中，所述通过所述NPU根据所述采样指示报文生成所述目标数据的采样报告报文，包括：
    在所述采样指示报文中所述抑制冗余标识为有效值的情况下，通过所述NPU根据所述采样指示报文，更新本地存储的采样值，并判断当前时刻的采样值是否发生变化；
    通过所述NPU响应于确定所述当前时刻的采样值发生变化，根据更新后的本地存储的采样值生成所述目标数据的采样报告报文；
    通过所述NPU响应于确定所述当前时刻的采样值未发生变化，在所述采样指示报文中的所述报告计数值等于所述报告间隔的情况下，根据更新后的本地存储的采样值生成所述目标数据的采样报告报文，并发送至CPU；
    其中，所述采样报告报文包括采样值报告报文和平均值报告报文中的至少之一。
  7. 一种网络设备的数据采样方法，应用于中央处理单元CPU，包括：
    将目标数据的采样参数配置至协处理器，其中，所述采样参数用于所述协处理器以设定频率生成采样指示报文并将所述采样指示报文发送至网络处理单元NPU，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至所述CPU；
    接收所述NPU发送的所述目标数据的采样报告报文。
  8. 根据权利要求7所述的方法，其中，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔；或者，所述采样参数至少包括：采样编号、采样功能项、采样频率、报告间隔、抑制冗余标识；
    其中，所述报告间隔包括：采样值报告间隔和平均值报告间隔中的至少之一，所述采样值报告间隔根据采样值报告频率和所述采样频率确定，所述平均值报告间隔根据平均值报告频率和所述采样频率确定。
  9. 一种网络设备的数据采样方法，应用于协处理器，包括：
    读取中央处理单元CPU配置的目标数据的采样参数；
    根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至网络处理单元NPU，其中，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至所述CPU。
  10. 根据权利要求9所述的方法，其中，所述根据所述采样参数以设定频率生成采样指示报文，包括：
    根据所述采样参数，对所述采样参数中的采样频率不为零的采样功能项，以不为零的所述采样频率生成采样指示报文；
    其中，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔；或者，所述采样指示报文中携带报文类型、采样编号、采样功能项、报告计数值、报告间隔、抑制冗余标识；
    在所述报告间隔包括采样值报告间隔的情况下，所述报告计数值包括采样值报告计数值；依次发送的采样指示报文中的采样值报告计数值由初始值依次递增至所述采样值报告间隔，在发送的采样指示报文中的采样值报告计数值为所述采样值报告间隔的情况下，重置下一次发送的采样指示报文中的采样值报告计数值为所述初始值；
    在所述报告间隔包括平均值报告间隔的情况下，所述报告计数值包括平均值报告计数值；对所述平均值报告间隔不为零的采样功能项，依次发送的采样指示报文中的平均值报告计数值由初始值依次递增至所述平均值报告间隔，在发送的采样指示报文中的平均值报告计数值为所述平均值报告间隔的情况下，重置下一次发送的采样指示报文中的平均值报告计数值为所述初始值。
  11. 根据权利要求10所述的方法，其中，所述采样指示报文中还包括：采样功能项总数量，所述采样功能项总数量用于指示所述采样指示报文中包括的采样编号以及采样功能项的组数量；
    不同组中的采样功能项的采样频率和报告间隔均相同。
  12. 一种网络设备的数据采样方法，应用于网络处理单元NPU，包括：
    接收协处理器发送的采样指示报文，其中，所述采样指示报文为所述协处理器根据中央处理单元CPU发送的目标数据的采样参数以设定频率生成并发送的；
    根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至所述CPU。
  13. 根据权利要求12所述的方法，其中，所述根据所述采样指示报文生成所述目标数据的采样报告报文，包括：
    根据所述采样指示报文，更新本地存储的采样值；
    在所述采样指示报文中的报告计数值等于报告间隔的情况下，根据更新后的本地存储的采样值生成所述目标数据的采样报告报文；
    其中，所述采样报告报文包括采样值报告报文和平均值报告报文中的至少之一。
  14. 根据权利要求12所述的方法，其中，根据所述采样指示报文生成所述目标数据的采样报告报文，包括：
    在所述采样指示报文中的抑制冗余标识为有效值的情况下，根据所述采样指示报文，更新本地存储的采样值，并判断当前时刻的采样值是否发生变化；
    响应于所述当前的时刻采样值发生变化，根据更新后的本地存储的采样值生成所述目标数据的采样报告报文；
    响应于所述当前的时刻采样值未发生变化，在所述采样指示报文中的报告计数值等于报告间隔的情况下，根据更新后的本地存储的采样值生成所述目标数据的采样报告报文；
    其中，所述采样报告报文包括采样值报告报文和平均值报告报文中的至少之一。
  15. 一种网络设备的数据采样装置，应用于中央处理单元CPU，包括：
    采样参数配置模块，设置为将目标数据的采样参数配置至协处理器，其中，所述采样参数用于所述协处理器以设定频率生成采样指示报文并将所述采样指示报文发送至网络处理单元NPU，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至所述CPU；
    采样报告报文接收模块，设置为接收所述NPU发送的所述目标数据的采样报告报文。
  16. 一种网络设备的数据采样装置，应用于协处理器，包括：
    采样参数读取模块，设置为读取中央处理单元CPU配置的目标数据的采样参数；
    采样指示报文发送模块，设置为根据所述采样参数以设定频率生成采样指示报文，并将所述采样指示报文发送至网络处理单元NPU，其中，所述采样指示报文用于所述NPU生成所述目标数据的采样报告报文并将所述采样报告报文发送至所述CPU。
  17. 一种网络设备的数据采样装置，应用于网络处理单元NPU，包括：
    采样指示报文接收模块，设置为接收协处理器发送的采样指示报文，其中，所述采样指示报文为所述协处理器根据中央处理单元CPU发送的目标数据的采样参数以设定频率生成并发送的；
    采样报告报文发送模块，设置为根据所述采样指示报文生成所述目标数据的采样报告报文，并将所述采样报告报文发送至所述CPU。
  18. 一种中央处理单元CPU，包括：
    至少一个处理单元；
    存储单元，设置为存储至少一个程序；
    当所述至少一个程序被所述至少一个处理单元执行，使得所述至少一个处理单元实现如权利要求7-8任一项所述的网络设备的数据采样方法。
  19. 一种协处理器，包括：
    至少一个处理单元；
    存储单元，设置为存储至少一个程序；
    当所述至少一个程序被所述至少一个处理单元执行，使得所述至少一个处理单元实现如权利要求9-11任一项所述的网络设备的数据采样方法。
  20. 一种网络处理单元NPU，包括：
    至少一个处理单元；
    存储单元，设置为存储至少一个程序；
    当所述至少一个程序被所述至少一个处理单元执行，使得所述至少一个处理单元实现如权利要求12-14任一项所述的网络设备的数据采样方法。
  21. 一种网络设备，包括：中央处理单元CPU、协处理器和网络处理单元NPU，
    其中，所述CPU执行如权利要求7-8任一项所述的网络设备的数据采样方法；
    所述协处理器执行如权利要求9-11任一项所述的网络设备的数据采样方法；
    所述NPU执行如权利要求12-14任一项所述的网络设备的数据采样方法。
  22. 根据权利要求21所述的网络设备，其中，所述协处理器包括外挂的现场可编程逻辑门阵列FPGA，或者所述NPU中的包发生器。
  23. 一种存储介质，存储有计算机程序，所述计算机程序被处理器执行时实 现权利要求1-6任一项所述的网络设备的数据采样方法，或者如权利要求7-8任一项所述的网络设备的数据采样方法，或者如权利要求9-11任一项所述的网络设备的数据采样方法，或者如权利要求12-14任一项所述的网络设备的数据采样方法。
PCT/CN2021/082475 2020-04-14 2021-03-23 网络设备的数据采样方法、装置、设备及介质 WO2021208682A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP21788665.4A EP4138340A4 (en) 2020-04-14 2021-03-23 DATA SAMPLING METHOD, APPARATUS AND DEVICE FOR NETWORK DEVICE AND MEDIUM
US17/914,670 US20230131524A1 (en) 2020-04-14 2021-03-23 Data sampling method for a network device, device, and medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010289203.2A CN113542043B (zh) 2020-04-14 2020-04-14 网络设备的数据采样方法、装置、设备及介质
CN202010289203.2 2020-04-14

Publications (1)

Publication Number Publication Date
WO2021208682A1 true WO2021208682A1 (zh) 2021-10-21

Family

ID=78084084

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/082475 WO2021208682A1 (zh) 2020-04-14 2021-03-23 网络设备的数据采样方法、装置、设备及介质

Country Status (4)

Country Link
US (1) US20230131524A1 (zh)
EP (1) EP4138340A4 (zh)
CN (1) CN113542043B (zh)
WO (1) WO2021208682A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114095383A (zh) * 2022-01-20 2022-02-25 紫光恒越技术有限公司 网络流量采样方法、系统和电子设备
CN114567574A (zh) * 2022-03-01 2022-05-31 烽火通信科技股份有限公司 一种基于时序控制实现lm免流量测试的方法与装置

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116662100B (zh) * 2022-09-27 2024-03-15 荣耀终端有限公司 数据处理方法和电子设备
CN118509043A (zh) * 2023-02-15 2024-08-16 华为技术有限公司 应用于故障定位的数据处理方法及光模块

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102546399A (zh) * 2011-12-16 2012-07-04 广东电网公司茂名供电局 一种智能变电站过程层报文线性处理架构及其处理方法
CN103200048A (zh) * 2013-04-02 2013-07-10 中兴通讯股份有限公司 一种网络处理器异常检测方法、装置及网络处理设备
CN103368777A (zh) * 2013-07-11 2013-10-23 曙光信息产业股份有限公司 一种数据包处理板及处理方法
CN104915181A (zh) * 2014-03-15 2015-09-16 英特尔公司 条件存储器错误帮助抑制
US20190087712A1 (en) * 2017-09-18 2019-03-21 Qualcomm Incorporated Neural Network Co-Processing

Family Cites Families (54)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3821709A (en) * 1972-10-05 1974-06-28 Honeywell Inf Systems Memory storage sequencer
US7917647B2 (en) * 2000-06-16 2011-03-29 Mcafee, Inc. Method and apparatus for rate limiting
US9438504B2 (en) * 2009-04-03 2016-09-06 Wisconsin Alumni Research Foundation Network routing system providing increased network bandwidth
WO2011066867A1 (en) * 2009-12-04 2011-06-09 Telefonaktiebolaget L M Ericsson (Publ) Random data stream sampling
CN101951024B (zh) * 2010-08-18 2013-04-10 国电南瑞科技股份有限公司 一种数字化数据采集装置
CN101976217B (zh) * 2010-10-29 2014-06-04 中兴通讯股份有限公司 网络处理器异常检测方法及系统
CN102346455B (zh) * 2011-04-29 2013-07-10 山东科汇电力自动化有限公司 采样模块的设计方法
US8958318B1 (en) * 2011-09-21 2015-02-17 Cisco Technology, Inc. Event-based capture of packets from a network flow
WO2015114646A1 (en) * 2014-01-30 2015-08-06 Hewlett-Packard Development Company, L.P. Analyzing network traffic in a computer network
US9306818B2 (en) * 2014-07-17 2016-04-05 Cellos Software Ltd Method for calculating statistic data of traffic flows in data network and probe thereof
US10044581B1 (en) * 2015-09-29 2018-08-07 Amazon Technologies, Inc. Network traffic tracking using encapsulation protocol
CN104202209B (zh) * 2014-08-05 2017-12-22 北京遥测技术研究所 一种基于cuda的多体制信号频率快速捕获实现方法
US20160088001A1 (en) * 2014-09-22 2016-03-24 Alcatel-Lucent Usa Inc. Collaborative deep packet inspection systems and methods
US10467036B2 (en) * 2014-09-30 2019-11-05 International Business Machines Corporation Dynamic metering adjustment for service management of computing platform
US10469342B2 (en) * 2014-10-10 2019-11-05 Nicira, Inc. Logical network traffic analysis
WO2016122708A1 (en) * 2015-01-28 2016-08-04 Hewlett Packard Enterprise Development Lp Determining a sampling rate for data traffic
US10108516B2 (en) * 2015-01-29 2018-10-23 Knuedge Incorporated Affinity data collection in a computing system
CN204462752U (zh) * 2015-03-22 2015-07-08 国家电网公司 一种信号采集处理系统
CN105323724B (zh) * 2015-06-03 2019-06-07 成都西加云杉科技有限公司 组播/广播技术中发送速率的设置方法及装置
US10356660B2 (en) * 2015-09-25 2019-07-16 Carlos Giraldo Rodriguez Systems and methods for optimizing network traffic
CN106921637B (zh) * 2015-12-28 2020-02-14 华为技术有限公司 网络流量中的应用信息的识别方法和装置
CN105610633A (zh) * 2016-02-23 2016-05-25 烽火通信科技股份有限公司 一种通信设备中实时性能自采样系统及方法
CN105763480B (zh) * 2016-04-12 2018-09-18 烽火通信科技股份有限公司 一种基于fpga实现流采样的装置及方法
CN109479015B (zh) * 2016-05-18 2022-09-09 马维尔以色列(M.I.S.L.)有限公司 网络设备及其延迟监测方法
US10785278B2 (en) * 2016-11-04 2020-09-22 Google Llc Network management interface
CN106970300B (zh) * 2017-03-09 2019-06-28 东北电力大学 基于神经网络处理器的小电流接地选线装置及其操控方法
US20180278500A1 (en) * 2017-03-24 2018-09-27 The Trustees Of Princeton University Scalable streaming analytics platform for network monitoring
CN109392002B (zh) * 2017-08-11 2022-07-12 华为技术有限公司 一种上报网络性能参数的方法及设备
US10972358B2 (en) * 2017-08-30 2021-04-06 Citrix Systems, Inc. Inferring congestion and signal quality
US10972397B2 (en) * 2017-09-29 2021-04-06 Futurewei Technologies, Inc. Self-driving packets with conditional commands
US10547560B1 (en) * 2017-12-28 2020-01-28 Juniper Networks, Inc. Monitoring network communications queues
CN108429657A (zh) * 2018-02-28 2018-08-21 新华三技术有限公司 报文采样方法、装置及报文处理设备
US11388074B2 (en) * 2018-04-12 2022-07-12 Intel Corporation Technologies for performance monitoring and management with empty polling
EP3804396A4 (en) * 2018-06-01 2022-03-02 Telefonaktiebolaget LM Ericsson (publ) METHOD FOR GATHERING SAMPLING DATA FOR WIRELESS TRANSMISSION
US11620528B2 (en) * 2018-06-12 2023-04-04 Ciena Corporation Pattern detection in time-series data
US10599548B2 (en) * 2018-06-28 2020-03-24 Intel Corporation Cache monitoring
WO2020000333A1 (en) * 2018-06-29 2020-01-02 SZ DJI Technology Co., Ltd. Image processing method and apparatus
CN109474362A (zh) * 2018-09-14 2019-03-15 许继集团有限公司 就地化保护采样数据自适应同步方法和就地化保护子机
US10644947B2 (en) * 2018-09-27 2020-05-05 International Business Machines Corporation Non-invasive diagnosis of configuration errors in distributed system
CN112703487B (zh) * 2018-10-26 2024-08-27 威睿有限责任公司 在数据中心中分层收集样本
US11443166B2 (en) * 2018-10-29 2022-09-13 Oracle International Corporation Datacenter level utilization prediction without operating system involvement
CN113348645B (zh) * 2018-11-27 2024-02-27 萨瑟尔公司 数据流分类的系统和方法
WO2020121294A1 (en) * 2018-12-10 2020-06-18 Drivenets Ltd. A system and a method for monitoring traffic flows in a communications network
CN109861981A (zh) * 2018-12-28 2019-06-07 中科院计算技术研究所南京移动通信与计算创新研究院 一种信号变频采样测试控制方法、系统、装置及存储介质
MX2021010315A (es) * 2019-02-28 2021-11-12 Assia Spe Llc Sistemas y metodos de administracion del espectro ergodico.
US11435937B2 (en) * 2019-03-26 2022-09-06 EMC IP Holding Company LLC Monitoring for service processors
US11533326B2 (en) * 2019-05-01 2022-12-20 Oracle International Corporation Systems and methods for multivariate anomaly detection in software monitoring
US11328222B1 (en) * 2019-05-10 2022-05-10 Innovium, Inc. Network switch with integrated gradient aggregation for distributed machine learning
CN110266726B (zh) * 2019-07-08 2021-07-20 新华三信息安全技术有限公司 一种识别ddos攻击数据流的方法及装置
US11240696B2 (en) * 2019-07-15 2022-02-01 Qualcomm Incorporated Measuring and verifying layer 2 sustained downlink maximum data rate decoding performance
US12067373B2 (en) * 2019-09-25 2024-08-20 Arm Limited Hybrid filter banks for artificial neural networks
US11710492B2 (en) * 2019-10-02 2023-07-25 Qualcomm Incorporated Speech encoding using a pre-encoded database
US11729116B2 (en) * 2020-03-10 2023-08-15 Dell Products L.P. Violation detection and isolation of endpoint devices in soft zoning environment
EP4162335A4 (en) * 2020-07-21 2024-01-24 Siemens Aktiengesellschaft DYNAMIC MULTIPARAMETER SCANNING METHOD AND DYNAMIC MULTIPARAMETER SCANNING APPARATUS

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102546399A (zh) * 2011-12-16 2012-07-04 广东电网公司茂名供电局 一种智能变电站过程层报文线性处理架构及其处理方法
CN103200048A (zh) * 2013-04-02 2013-07-10 中兴通讯股份有限公司 一种网络处理器异常检测方法、装置及网络处理设备
CN103368777A (zh) * 2013-07-11 2013-10-23 曙光信息产业股份有限公司 一种数据包处理板及处理方法
CN104915181A (zh) * 2014-03-15 2015-09-16 英特尔公司 条件存储器错误帮助抑制
US20190087712A1 (en) * 2017-09-18 2019-03-21 Qualcomm Incorporated Neural Network Co-Processing

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4138340A4 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114095383A (zh) * 2022-01-20 2022-02-25 紫光恒越技术有限公司 网络流量采样方法、系统和电子设备
CN114095383B (zh) * 2022-01-20 2022-04-12 紫光恒越技术有限公司 网络流量采样方法、系统和电子设备
CN114567574A (zh) * 2022-03-01 2022-05-31 烽火通信科技股份有限公司 一种基于时序控制实现lm免流量测试的方法与装置
CN114567574B (zh) * 2022-03-01 2023-11-10 烽火通信科技股份有限公司 一种基于时序控制实现lm免流量测试的方法与装置

Also Published As

Publication number Publication date
CN113542043A (zh) 2021-10-22
CN113542043B (zh) 2024-06-07
US20230131524A1 (en) 2023-04-27
EP4138340A1 (en) 2023-02-22
EP4138340A4 (en) 2024-04-17

Similar Documents

Publication Publication Date Title
WO2021208682A1 (zh) 网络设备的数据采样方法、装置、设备及介质
US10812333B2 (en) Microservice configuration apparatus and method
CN110222091B (zh) 一种海量数据实时统计分析方法
JP7039685B2 (ja) トラフィック測定方法、デバイス、およびシステム
WO2021259013A1 (zh) 数据处理方法、装置、电子设备及计算机可读介质
US20170091013A1 (en) Pcie error reporting and throttling
CN108023829B (zh) 报文处理方法及装置、存储介质、电子设备
WO2019232927A1 (zh) 分布式数据删除流控方法、装置、电子设备及存储介质
US20210058811A1 (en) Method and apparatus for detecting time series data
CN111290696A (zh) 一种应用程序组件的流控方法及装置
CN111949568A (zh) 一种报文处理方法、装置及网络芯片
CN108206787A (zh) 一种拥塞避免方法和装置
US11411799B2 (en) Scalable statistics and analytics mechanisms in cloud networking
US11818036B2 (en) Data processing method, apparatus, and device
US9350678B2 (en) Controller based network resource management
CN117997802B (zh) 一种网络抖动检测方法、装置、设备及介质
WO2019109902A1 (zh) 队列调度方法及装置、通信设备、存储介质
CN114020529A (zh) 一种流表数据的备份方法、装置、网络设备和存储介质
US10185675B1 (en) Device with multiple interrupt reporting modes
CN117097679A (zh) 一种网络中断的聚合方法、装置及网络通信设备
WO2016082614A1 (zh) 一种组播路由表项处理方法、装置及计算机存储介质
US10560317B2 (en) Subscription to a subset of switching events
WO2017016300A1 (zh) 令牌申请处理方法、装置和计算机存储介质
US9674282B2 (en) Synchronizing SLM statuses of a plurality of appliances in a cluster
CN112711549B (zh) 中断请求信号转换系统和方法、计算装置

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: 21788665

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2021788665

Country of ref document: EP

Effective date: 20221114