Service flow analysis method, apparatus, and system

Abstract

The present application relates to the technical field of networks, and provides a service flow analysis method, apparatus, and system. A network management device acquires measurement task information of a service, and issues measurement commands to a plurality of network devices, so as to instruct the network devices to count traffic characteristics of flows passing through the network devices. The service comprises a plurality of service flows, and the measurement task information comprises device identifiers of a plurality of terminal devices associated with the service. After receiving measurement results reported by the plurality of network devices, on the basis of the measurement results sent by the plurality of network devices and the device identifiers of the plurality of terminal devices, the network management device determines a traffic analysis result corresponding to the service. According to the present application, traffic characteristics corresponding to flow identifiers of flows passing through a plurality of network devices in a network are collected, so that a traffic analysis result corresponding to a service can be determined on the basis of the traffic characteristics of the plurality of service flows of the service, thereby achieving network measurement at the granularity of services, and facilitating monitoring and analysis of the service quality.

Description

Business flow analysis method, device and system

[0001] This application claims priority to Chinese patent application number 202311820481.6, filed on December 26, 2023, entitled “Business Traffic Analysis Method, Device and System”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of network technology, and in particular to a business traffic analysis method, device and system. Background Art

[0003] With the rapid development of network technology, more and more services require network traffic transmission during operation. Since the transmission of service traffic in the network directly affects the service quality, it is crucial to analyze the transmission of service traffic in the network. Summary of the Invention

[0004] The present application provides a business traffic analysis method, device and system.

[0005] In a first aspect, a service traffic analysis method is provided. The method can be applied to a network management device. The method includes: the network management device obtains measurement task information for a service, where the measurement task information includes device identifiers of multiple terminal devices associated with the service. The service includes multiple service flows, and the multiple terminal devices include sending devices and receiving devices of the multiple service flows. The network management device sends a measurement command to multiple network devices in the network, where the measurement command is used to instruct the network device that received the measurement command to count traffic characteristics of one or more flows flowing through the network device. The network management device receives measurement results sent by the multiple network devices. The measurement results sent by each network device include traffic characteristics corresponding to the flow identifiers of one or more flows flowing through the network device. The flow identifier of each flow includes the device identifier of the sending device of the flow and the device identifier of the receiving device of the flow. The network management device determines a traffic analysis result corresponding to the service based on the measurement results sent by the multiple network devices and the device identifiers of the multiple terminal devices.

[0006] The present application collects the traffic characteristics corresponding to the flow identifiers of flows passing through multiple network devices in the network through a network management device, so that the network management device can determine the multiple business flows of a business with multiple business flows based on the collected flow identifiers and the device identifier in the measurement task information of the business, and then determine the traffic analysis results corresponding to the business based on the traffic characteristics of the multiple business flows, thereby realizing network measurement with business as the granularity, which is helpful for monitoring and analyzing business quality.

[0007] Optionally, the service is a collective communication or a distributed training task, wherein the distributed training task may include one or more collective communications.

[0008] This application can analyze the traffic characteristics of each business flow in the collective communication based on the characteristics of the collective communication, and present the traffic analysis results at the granularity of the collective communication, so as to coordinate with the computing tasks and assist in analyzing whether the slowness or failure of distributed training is caused by the network side or the computing side.

[0009] Optionally, the network management device determines the traffic analysis result corresponding to the service based on the measurement results sent by multiple network devices and the device identifiers of multiple terminal devices, including: for each measurement result sent by the network device, the network management device determines the service flow belonging to the service from one or more flows flowing through the network device based on the flow identifier in the measurement result and the device identifiers of the multiple terminal devices. The network management device determines the traffic analysis result based on the traffic characteristics of the multiple service flows in the measurement results sent by the multiple network devices. The network management device may determine a flow whose sending device and receiving device both belong to the multiple terminal devices as the service flow of the service.

[0010] In a first embodiment, multiple network devices include a first network device, and the measurement results sent by the first network device include a first traffic characteristic of a first business flow, the first business flow is any business flow among multiple business flows, and the first business flow includes multiple message groups used to complete multiple data transmissions. The first traffic characteristic includes the transmission start time and transmission end time of each of the multiple message groups. The network management device determines the implementation method of the traffic analysis result based on the traffic characteristics of the multiple business flows in the measurement results sent by the multiple network devices, including: the network management device determines the flow completion time (FCT) of the first business flow within the first statistical period based on the transmission duration of one or more message groups with transmission end times within the first statistical period among the multiple message groups. The transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group. Among them, the traffic analysis result includes the flow completion time of the multiple business flows in multiple statistical periods, and the first statistical period is any statistical period among the multiple statistical periods.

[0011] Under this implementation method, the network device only needs to count the start and end times of each data transmission in the flow passing through itself. The network management device performs traffic analysis on the service based on the traffic characteristics reported by the network device to determine the flow completion time of multiple business flows of the service in multiple statistical time periods. This implementation method has low requirements on the processing performance of the network device.

[0012] In a second embodiment, the multiple network devices include a second network device, and the measurement results sent by the second network device include a second traffic characteristic of a second business flow, the second business flow is any business flow among the multiple business flows, and the second business flow includes multiple message groups respectively used to complete multiple data transmissions. The second traffic characteristic includes the flow completion time of the second business flow within the second statistical period. The flow completion time of the second business flow within the second statistical period is obtained based on the transmission duration of one or more message groups in the multiple message groups within the second statistical period at the end of the transmission. The transmission duration of each message group is obtained based on the transmission end time and the transmission start time of the message group. Among them, the traffic analysis results include the flow completion time of the multiple business flows in multiple statistical periods, and the second statistical period is any statistical period among the multiple statistical periods.

[0013] Under this implementation method, the network device counts the start and end times of each data transmission in the flow passing through itself, and performs traffic analysis to determine the flow completion time of the flow passing through itself within multiple statistical time periods. The network management device summarizes and counts the traffic characteristics reported by multiple network devices, and can obtain the flow completion time of multiple business flows of the business within multiple statistical time periods. This implementation method has low processing performance requirements for the network management device and can improve the efficiency of the network management device in analyzing the traffic of the business.

[0014] Optionally, the flow completion time of a service flow within a statistical period is a statistical value of the transmission duration of all message groups in the service flow whose transmission ends within the statistical period. The statistical value can be an average, median, maximum, or percentile value.

[0015] Optionally, for any of the multiple service flows, if the difference between the flow completion time of the service flow in the third statistical period and the flow completion time of the service flow in the fourth statistical period is greater than a first threshold, the network management device determines that the transmission quality of the service flow has deteriorated. The third statistical period is chronologically located after the fourth statistical period. In other words, if the flow completion time of a service flow becomes longer, it indicates that the transmission quality of the service flow has deteriorated. The traffic analysis results also include an indication of the service flow among the multiple service flows whose transmission quality has deteriorated.

[0016] Under the above-mentioned first or second implementation mode, by measuring the flow completion time of multiple business flows of the same business in multiple statistical time periods, the changes in the flow completion time of multiple business flows over time are presented with the business as the granularity, wherein the changes in the flow completion time of a single business flow over time can reflect the changes in the transmission quality of the business flow itself, and the changes in the flow completion time of multiple business flows over time can reflect the changes in the overall transmission quality of the business. This helps operation and maintenance personnel analyze whether the overall transmission quality of the business has deteriorated and the specific business flows that cause the degradation of the business transmission quality.

[0017] In a third embodiment, the multiple network devices include a third network device, and the measurement results sent by the third network device include third traffic characteristics of a third service flow, where the third service flow is any service flow from among the multiple service flows, and the third service flow includes multiple message groups used to complete multiple data transmissions. The third traffic characteristics include the transmission start time, transmission end time, and traffic volume of each of the multiple message groups. A network management device determines, based on the traffic characteristics of the multiple service flows in the measurement results sent by the multiple network devices, a method for implementing the traffic analysis results, including: the network management device determines the throughput of the third service flow by the third network device within a fifth statistical period based on the transmission duration and traffic volume of one or more message groups from the multiple message groups whose transmission end time falls within a fifth statistical period. The transmission duration of each message group is determined based on the transmission end time and transmission start time of the message group. The traffic analysis results include the throughput of the service flows flowing through each of the multiple network devices within multiple statistical periods, where the fifth statistical period is any one of the multiple statistical periods.

[0018] Under this implementation method, the network device only needs to count the start time, end time and flow size of each data transmission in the flow passing through itself. The network management device performs traffic analysis on the service based on the traffic characteristics reported by multiple network devices to determine the throughput of multiple service flows of the service by multiple network devices in multiple statistical time periods. This implementation method has low requirements on the processing performance of the network device.

[0019] In a fourth embodiment, the multiple network devices include a fourth network device, and the measurement results sent by the fourth network device include the fourth traffic characteristics of the fourth business flow, the fourth business flow is any business flow among the multiple business flows, and the fourth business flow includes multiple message groups respectively used to complete multiple data transmissions. The fourth traffic characteristics include the throughput of the fourth business flow by the fourth network device within the sixth statistical period, and the throughput of the fourth business flow within the sixth statistical period is obtained based on the transmission duration and traffic size of one or more message groups among the multiple message groups whose transmission ends within the sixth statistical period. The transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group. Among them, the traffic analysis results include the throughput of the business flows flowing through the multiple network devices in multiple statistical periods, and the sixth statistical period is any statistical period among the multiple statistical periods.

[0020] Under this implementation, the network device counts the start time, end time and flow size of each data transmission in the flow passing through itself, and performs traffic analysis to determine the throughput of the flow passing through itself in multiple statistical time periods. The network management device summarizes and counts the traffic characteristics reported by multiple network devices, and can obtain the throughput of multiple business flows of the business for multiple network devices in multiple statistical time periods. This implementation method has low processing performance requirements for the network management device and can improve the efficiency of the network management device's traffic analysis of the business.

[0021] Optionally, for any of the multiple service flows, if the difference between the throughput of any network device through which the service flow passes for the service flow during the seventh statistical period and the throughput of the network device for the service flow during the eighth statistical period is greater than a second threshold, the network management device determines that the transmission quality of the service flow has deteriorated. The eighth statistical period is chronologically located after the seventh statistical period. In other words, if the throughput of the network device for a particular service flow decreases, it indicates that the transmission quality of the service flow has deteriorated. The traffic analysis results also include an indication of the service flow among the multiple service flows whose transmission quality has deteriorated.

[0022] Under the above-mentioned third or fourth implementation mode, by measuring the throughput of multiple business flows of the same business of the network device in multiple statistical time periods, the change of the throughput of the network device for the business flow over time is presented with the business as the granularity, wherein the change of the throughput of the network device for a single business flow over time can reflect the change of the transmission quality of the business flow itself, and the change of the throughput of multiple network devices for multiple business flows over time can reflect the change of the overall transmission quality of the business. This helps operation and maintenance personnel to analyze whether the overall transmission quality of the business has deteriorated and the specific business flow that causes the degradation of the business transmission quality.

[0023] Optionally, in combination with any one of the first to fourth embodiments mentioned above, the implementation method of the network management device sending measurement commands to multiple network devices in the network includes: the network management device sends measurement commands to multiple access network devices in the network respectively, and each access network device is used to connect one or more terminal devices among the multiple terminal devices to the network.

[0024] Because service flows sent or received by terminal devices associated with a service inevitably pass through access network devices, and the completion time and throughput of a service flow across the network are essentially the same across different network devices, when the traffic characteristics being counted include completion time and / or throughput, only the traffic characteristics of flows passing through the access network devices are required. This reduces the number of network devices involved in service traffic analysis, while ensuring that the traffic characteristics of all service flows are counted, conserving communication resources between network devices and network management devices, as well as processing resources on each device.

[0025] In a fifth implementation mode, the measurement results sent by each network device respectively include the flow size of one or more flows flowing through the network device. The network management device determines the implementation method of the flow analysis result based on the flow characteristics of multiple business flows in the measurement results sent by multiple network devices, including: for each network device, the network management device determines the cumulative flow size of the network device for the business based on the flow size of all business flows in the measurement results sent by the network device. The network management device determines the load balancing degree of the multiple network devices for the business based on the cumulative flow size of the multiple network devices for the business. The load balancing degree is negatively correlated with the difference between the cumulative flow sizes of the multiple network devices for the business. Among them, the flow analysis result includes the load balancing degree of the multiple network devices for the business.

[0026] The higher the load balancing degree, the more evenly the traffic distribution on the network device is when the multiple business flows of the service are transmitted in the network, and the lower the probability of congestion on the network device. The network management device of this application analyzes the load balancing degree of different network devices on the network path through which the business flow passes, which helps the operation and maintenance side analyze and adjust the transmission path of the business flow in the business, thereby improving the load balancing degree of multiple network devices for the business.

[0027] Optionally, in conjunction with the fifth embodiment described above, the network management device may send measurement commands to multiple network devices in the network by sending the measurement commands to all network devices along the network paths of the multiple service flows. This reduces the number of network devices involved in service flow analysis, thereby conserving communication resources between the network devices and the network management device, as well as processing resources of each network device.

[0028] Optionally, the measurement command also includes a measurement indication, which includes one or more of the following: a source Internet Protocol (IP) address set, the source IP address set includes the IP addresses of one or more terminal devices among a plurality of terminal devices, and the source IP address set is used to instruct the network device that receives the measurement command to count the traffic characteristics of flows passing through the network device and whose source IP addresses belong to the source IP address set; a destination IP address set, the destination IP address set includes the IP addresses of one or more terminal devices among a plurality of terminal devices, and the destination IP address set is used to instruct the network device that receives the measurement command to count the traffic characteristics of flows passing through the network device and whose destination IP addresses belong to the destination IP address set; an inbound direction indication, the inbound direction indication is used to instruct the network device that receives the measurement command to count the traffic characteristics of one or more flows received by the network device; an outbound direction indication, the outbound direction indication is used to instruct the network device that receives the measurement command to count the traffic characteristics of one or more flows sent by the network device; an interface indication, the interface indication is used to instruct the network device that receives the measurement command to count the traffic characteristics of flows passing through one or more specified interfaces of the network device, and the measurement result sent by the network device also includes an interface identifier, and the interface identifier is used to indicate the interface through which the flow passes on the network device.

[0029] In this application, the network management device carries measurement instructions in the measurement command, so that the network device can perform targeted statistics on the traffic characteristics of some flows, reduce the network device's statistics on the traffic characteristics of useless flows, and thus save the processing resources of the network device.

[0030] Optionally, the measurement task information also includes a traffic analysis task, which is used to indicate a traffic analysis type. The network management device may generate a measurement command based on the traffic analysis task, where the type of traffic characteristics to be statistically analyzed indicated by the measurement command matches the traffic analysis type indicated by the traffic analysis task. Accordingly, the network management device determines an implementation method for the traffic analysis result corresponding to the service based on the measurement results sent by multiple network devices and the device identifiers of multiple terminal devices, including: the network management device determines the traffic analysis result of the service under the traffic analysis type indicated by the traffic analysis task based on the measurement results sent by multiple network devices and the device identifiers of multiple terminal devices.

[0031] Optionally, an implementation manner in which the network management device obtains the measurement task information of a service includes: the network management device receives a service measurement task sent by a service platform, where the service measurement task includes the measurement task information.

[0032] Optionally, the network management device sends a service measurement result to the service platform, where the service measurement result includes a traffic analysis result.

[0033] Optionally, the above-mentioned multiple business flows are remote direct memory access (RDMA) flows.

[0034] In a second aspect, a service traffic analysis method is provided. This method can be applied to a service platform. The method comprises: the service platform sends a service measurement task to a network management device. The service measurement task includes measurement task information for the service. The measurement task information includes device identifiers of multiple terminal devices associated with the service. The service includes multiple service flows, and the multiple terminal devices include sending devices and receiving devices for the multiple service flows. The service platform receives the service measurement results sent by the network management device. The service measurement results include traffic analysis results corresponding to the service.

[0035] In this application, the service platform can send a service measurement task to the network management device to trigger the network management device to perform a service traffic analysis process. The service platform can then receive the traffic analysis results sent by the network management device, allowing operation and maintenance personnel to view the running status of the service in the network through the service platform.

[0036] Optionally, the above-mentioned business is a collective communication or distributed training task.

[0037] Optionally, the service platform outputs the traffic analysis result. The service platform may display the traffic analysis result, or the service platform may send the traffic analysis result to a display device for display, so that operation and maintenance personnel can view the traffic analysis result corresponding to the service to achieve quality monitoring and management of the service.

[0038] In a third aspect, a service traffic analysis method is provided. This method can be applied to a network device. The method includes: the network device receives a measurement command sent by a network management device, the measurement command being used to instruct the network device to collect traffic characteristics of one or more flows passing through the network device. The network device collects traffic characteristics of the one or more flows passing through the network device according to the measurement command. The network device sends a measurement result to the network management device, the measurement result including traffic characteristics corresponding to flow identifiers of the one or more flows passing through the network device. The flow identifier of each flow includes the device identifier of the sending device of the flow and the device identifier of the receiving device of the flow.

[0039] Optionally, the measurement command also includes a measurement indication, which includes one or more of the following: a source IP address set, which includes one or more IP addresses, and the source IP address set is used to instruct the network device to count the traffic characteristics of flows passing through the network device and whose source IP addresses belong to the source IP address set; a destination IP address set, which includes one or more IP addresses, and the destination IP address set is used to instruct the network device to count the traffic characteristics of flows passing through the network device and whose destination IP addresses belong to the destination IP address set; an inbound direction indication, which is used to instruct the network device to count the traffic characteristics of one or more flows received by the network device; an outbound direction indication, which is used to instruct the network device to count the traffic characteristics of one or more flows sent by the network device; an interface indication, which is used to instruct the network device to count the traffic characteristics of flows passing through one or more specified interfaces of the network device. The measurement result sent by the network device also includes an interface identifier, which is used to indicate the interface through which the flow passes on the network device.

[0040] In a first embodiment, the measurement results include a first traffic characteristic of a first flow, where the first flow is any one of one or more flows flowing through the network device, and the first flow includes multiple message groups respectively used to complete multiple data transmissions, and the first traffic characteristic includes the transmission start time and transmission end time of each of the multiple message groups.

[0041] In a second embodiment, the measurement result includes a second traffic characteristic of a second flow, where the second flow is any one of one or more flows flowing through the network device, and the second flow includes multiple message groups used to complete multiple data transmissions. The second traffic characteristic includes a flow completion time of the second flow within one or more statistical time periods. The flow completion time of the second flow within each statistical time period is obtained based on the transmission duration of one or more message groups among the multiple message groups whose transmission ends within the statistical time period. The transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group.

[0042] In a third embodiment, the measurement results include a third traffic characteristic of a third flow, where the third flow is any one of one or more flows flowing through the network device, and the third flow includes multiple message groups respectively used to complete multiple data transmissions, and the third traffic characteristic includes the transmission start time, transmission end time and traffic size of each of the multiple message groups.

[0043] In a fourth embodiment, the measurement result includes a fourth traffic characteristic of a fourth flow, where the fourth flow is any one of one or more flows flowing through the network device, and the fourth flow includes multiple message groups respectively used to complete multiple data transmissions. The fourth traffic characteristic includes the throughput of the network device for the fourth flow within one or more statistical time periods. The throughput of the fourth flow within each statistical time period is obtained based on the transmission duration and flow size of one or more message groups among the multiple message groups at the end of transmission within the statistical time period. The transmission duration of each message group is obtained based on the end time and start time of transmission of the message group.

[0044] In a fifth implementation manner, the measurement result includes the flow rate of one or more flows passing through the network device.

[0045] In a fourth aspect, a service traffic analysis device is provided. This device can be applied to a network management device and includes multiple functional modules that interact with each other to implement the method described in the first aspect and its respective embodiments. The multiple functional modules can be implemented based on software, hardware, or a combination of software and hardware, and the multiple functional modules can be arbitrarily combined or divided based on the specific implementation.

[0046] For example, the device includes but is not limited to an acquisition module, a sending module, a receiving module, and a determination module. Optionally, the device also includes a generation module.

[0047] Among them, the acquisition module is used to obtain the measurement task information of the service, the measurement task information includes the device identification of multiple terminal devices associated with the service, the service includes multiple service flows, and the multiple terminal devices include the sending device and the receiving device of the multiple service flows. The sending module is used to send a measurement command to multiple network devices in the network, and the measurement command is used to instruct the network device that receives the measurement command to count the traffic characteristics of one or more flows flowing through the network device. The receiving module is used to receive the measurement results sent by the multiple network devices, and the measurement results sent by each network device respectively include the traffic characteristics corresponding to the flow identification of one or more flows flowing through the network device, and the flow identification of each flow includes the device identification of the sending device of the flow and the device identification of the receiving device of the flow. The determination module is used to determine the traffic analysis result corresponding to the service based on the measurement results sent by the multiple network devices and the device identification of the multiple terminal devices.

[0048] Optionally, the service is a collective communication or distributed training task.

[0049] Optionally, the determination module is used to: for the measurement results sent by each network device, determine the business flow belonging to the business in one or more flows flowing through the network device based on the flow identifier in the measurement results and the device identifiers of the multiple terminal devices; and determine the traffic analysis result based on the traffic characteristics of the multiple business flows in the measurement results sent by the multiple network devices.

[0050] In a first embodiment, the multiple network devices include a first network device, and the measurement results sent by the first network device include a first traffic characteristic of a first business flow, the first business flow is any business flow among the multiple business flows, and the first business flow includes multiple message groups respectively used to complete multiple data transmissions. The first traffic characteristic includes the transmission start time and transmission end time of each of the multiple message groups, and the determination module is used to: determine the flow completion time of the first business flow within the first statistical period based on the transmission duration of one or more message groups with transmission end times within the first statistical period among the multiple message groups, and the transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group. Wherein, the traffic analysis result includes the flow completion time of the multiple business flows in multiple statistical periods respectively, and the first statistical period is any statistical period among the multiple statistical periods.

[0051] In a second embodiment, the multiple network devices include a second network device, and the measurement results sent by the second network device include a second traffic characteristic of a second business flow, the second business flow is any business flow among the multiple business flows, and the second business flow includes multiple message groups respectively used to complete multiple data transmissions. The second traffic characteristic includes the flow completion time of the second business flow within the second statistical period, and the flow completion time of the second business flow within the second statistical period is obtained based on the transmission duration of one or more message groups among the multiple message groups whose transmission end moments are within the second statistical period, and the transmission duration of each message group is obtained based on the transmission end moment and transmission start moment of the message group. Wherein, the traffic analysis result includes the flow completion time of the multiple business flows in multiple statistical periods respectively, and the second statistical period is any statistical period among the multiple statistical periods.

[0052] Optionally, the flow completion time of a service flow in a statistical period is a statistical value of the transmission duration of all message groups in the service flow within the statistical period at the transmission end time.

[0053] Optionally, in combination with the first or second embodiment described above, the determination module is further configured to determine, for any one of the multiple service flows, that the transmission quality of the service flow has deteriorated if a difference between a flow completion time of the service flow within a third statistical period and a flow completion time of the service flow within a fourth statistical period is greater than a first threshold, and the third statistical period is chronologically subsequent to the fourth statistical period. The traffic analysis result further includes an indication of the service flow among the multiple service flows whose transmission quality has deteriorated.

[0054] In a third embodiment, the multiple network devices include a third network device, and the measurement results sent by the third network device include a third traffic characteristic of a third business flow, wherein the third business flow is any business flow among the multiple business flows, and the third business flow includes multiple message groups respectively used to complete multiple data transmissions. The third traffic characteristic includes the transmission start time, transmission end time, and traffic size of each of the multiple message groups, and the determination module is used to: determine the throughput of the third business flow of the third network device within the fifth statistical period based on the transmission duration and traffic size of one or more message groups among the multiple message groups whose transmission end time is within the fifth statistical period, and the transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group. The traffic analysis results include the throughput of the business flows flowing through the multiple network devices in multiple statistical periods, and the fifth statistical period is any statistical period among the multiple statistical periods.

[0055] In a fourth embodiment, the multiple network devices include a fourth network device, and the measurement results sent by the fourth network device include the fourth traffic characteristics of the fourth business flow, the fourth business flow is any business flow among the multiple business flows, and the fourth business flow includes multiple message groups respectively used to complete multiple data transmissions. The fourth traffic characteristics include the throughput of the fourth business flow by the fourth network device within the sixth statistical period, and the throughput of the fourth business flow within the sixth statistical period is obtained based on the transmission duration and traffic size of one or more message groups in the multiple message groups within the sixth statistical period at the end of transmission, and the transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group. Wherein, the traffic analysis results include the throughput of the business flows flowing through the multiple network devices in multiple statistical periods, and the sixth statistical period is any statistical period among the multiple statistical periods.

[0056] Optionally, in combination with the third or fourth embodiment described above, the determination module is further configured to determine, for any of the multiple service flows, that the transmission quality of the service flow has deteriorated if the difference between the throughput of any network device through which the service flow passes for the service flow in a seventh statistical period and the throughput of the network device for the service flow in an eighth statistical period is greater than a second threshold, and the eighth statistical period is chronologically subsequent to the seventh statistical period. The traffic analysis result further includes an indication of the service flow among the multiple service flows whose transmission quality has deteriorated.

[0057] Optionally, in combination with any one of the first to fourth embodiments above, the sending module is used to: send the measurement command to multiple access network devices in the network respectively, and each access network device is used to connect one or more terminal devices among the multiple terminal devices to the network.

[0058] In a fifth embodiment, the measurement results sent by each network device include the flow size of one or more flows flowing through the network device, and the determination module is used to: for each network device, determine the cumulative flow size of the network device for the service based on the flow size of all service flows in the measurement results sent by the network device; and determine the load balancing degree of the multiple network devices for the service based on the cumulative flow size of the multiple network devices for the service, wherein the load balancing degree is negatively correlated with the difference between the cumulative flow sizes of the multiple network devices for the service. The traffic analysis result includes the load balancing degree of the multiple network devices for the service.

[0059] Optionally, in combination with the fifth implementation, the sending module is configured to send the measurement command to all network devices on the network path of the multiple service flows in the network respectively.

[0060] Optionally, the measurement command also includes a measurement indication, and the measurement indication includes one or more of the following: a source IP address set, the source IP address set includes the IP addresses of one or more terminal devices among the multiple terminal devices, and the source IP address set is used to instruct the network device that receives the measurement command to count the traffic characteristics of the flow passing through the network device, and the source IP address belongs to the source IP address set; a destination IP address set, the destination IP address set includes the IP addresses of one or more terminal devices among the multiple terminal devices, and the destination IP address set is used to instruct the network device that receives the measurement command to count the traffic characteristics of the flow passing through the network device, and the destination IP address The traffic characteristics of flows belonging to the set of destination IP addresses; an inbound direction indication, wherein the inbound direction indication is used to instruct the network device that has received the measurement command to count the traffic characteristics of one or more flows received by the network device; an outbound direction indication, wherein the outbound direction indication is used to instruct the network device that has received the measurement command to count the traffic characteristics of one or more flows sent by the network device; an interface indication, wherein the interface indication is used to instruct the network device that has received the measurement command to count the traffic characteristics of flows passing through one or more designated interfaces of the network device, and the measurement result sent by the network device also includes an interface identifier, and the interface identifier is used to indicate the interface through which the flow passes on the network device.

[0061] Optionally, the measurement task information also includes a traffic analysis task, where the traffic analysis task indicates a traffic analysis type. A generation module is configured to generate the measurement command based on the traffic analysis task, where the type of traffic characteristics to be statistically analyzed indicated by the measurement command matches the traffic analysis type indicated by the traffic analysis task. Accordingly, a determination module is configured to determine a traffic analysis result for the service under the traffic analysis type indicated by the traffic analysis task based on the measurement results sent by the multiple network devices and the device identifiers of the multiple terminal devices.

[0062] Optionally, the acquisition module is configured to receive a service measurement task sent by a service platform, where the service measurement task includes the measurement task information.

[0063] Optionally, the sending module is further configured to send a service measurement result to the service platform, where the service measurement result includes the traffic analysis result.

[0064] Optionally, the multiple business flows are RDMA flows.

[0065] In a fifth aspect, a service traffic analysis device is provided. This device can be applied to a service platform and includes multiple functional modules that interact with each other to implement the method described in the second aspect and its respective embodiments. The multiple functional modules can be implemented based on software, hardware, or a combination of software and hardware, and the multiple functional modules can be arbitrarily combined or divided based on the specific implementation.

[0066] For example, the device includes but is not limited to a sending module and a receiving module. Optionally, the device also includes an output module.

[0067] The sending module is configured to send a service measurement task to a network management device, the service measurement task including measurement task information for the service, the measurement task information including device identifiers of multiple terminal devices associated with the service, the service including multiple service flows, and the multiple terminal devices including sending devices and receiving devices for the multiple service flows. The receiving module is configured to receive service measurement results sent by the network management device, the service measurement results including traffic analysis results corresponding to the service.

[0068] Optionally, the service is a collective communication or distributed training task.

[0069] Optionally, an output module is used to output the traffic analysis result.

[0070] In a sixth aspect, a service traffic analysis device is provided. This device can be applied to a network device and includes multiple functional modules that interact with each other to implement the method described in the third aspect and its respective embodiments. The multiple functional modules can be implemented based on software, hardware, or a combination of software and hardware, and the multiple functional modules can be arbitrarily combined or divided based on the specific implementation.

[0071] For example, the device includes but is not limited to a receiving module, a processing module, and a sending module.

[0072] The receiving module is configured to receive a measurement command sent by a network management device, the measurement command being configured to instruct the network device to collect statistics on the traffic characteristics of one or more flows passing through the network device. The processing module is configured to collect statistics on the traffic characteristics of one or more flows passing through the network device according to the measurement command. The sending module is configured to send measurement results to the network management device, the measurement results including the traffic characteristics corresponding to the flow identifiers of the one or more flows passing through the network device, wherein the flow identifier of each flow includes the device identifier of the sending device of the flow and the device identifier of the receiving device of the flow.

[0073] Optionally, the measurement command also includes a measurement indication, and the measurement indication includes one or more of the following: a source IP address set, the source IP address set includes one or more IP addresses, and the source IP address set is used to instruct the network device to count the traffic characteristics of flows passing through the network device and whose source IP addresses belong to the source IP address set; a destination IP address set, the destination IP address set includes one or more IP addresses, and the destination IP address set is used to instruct the network device to count the traffic characteristics of flows passing through the network device and whose destination IP addresses belong to the destination IP address set; an inbound direction indication, the inbound direction indication is used to instruct the network device to count the traffic characteristics of one or more flows received by the network device; an outbound direction indication, the outbound direction indication is used to instruct the network device to count the traffic characteristics of one or more flows sent by the network device; an interface indication, the interface indication is used to instruct the network device to count the traffic characteristics of flows passing through one or more specified interfaces of the network device, and the measurement result sent by the network device also includes an interface identifier, and the interface identifier is used to indicate the interface through which the flow passes on the network device.

[0074] Optionally, the measurement result includes a first traffic characteristic of a first flow, where the first flow is any one of the one or more flows, the first flow includes multiple message groups respectively used to complete multiple data transmissions, and the first traffic characteristic includes the transmission start time and transmission end time of each of the multiple message groups.

[0075] Optionally, the measurement result includes a second traffic characteristic of a second flow, where the second flow is any one of the one or more flows, the second flow includes multiple message groups respectively used to complete multiple data transmissions, the second traffic characteristic includes the flow completion time of the second flow within one or more statistical time periods, the flow completion time of the second flow within each statistical time period is obtained based on the transmission end time of the multiple message groups and the transmission duration of one or more message groups within the statistical time period, and the transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group.

[0076] Optionally, the measurement result includes a third traffic characteristic of a third flow, where the third flow is any one of the one or more flows, and the third flow includes multiple message groups respectively used to complete multiple data transmissions, and the third traffic characteristic includes the transmission start time, transmission end time and traffic size of each of the multiple message groups.

[0077] Optionally, the measurement result includes a fourth traffic characteristic of a fourth flow, where the fourth flow is any one of the one or more flows, and the fourth flow includes multiple message groups respectively used to complete multiple data transmissions. The fourth traffic characteristic includes the throughput of the network device for the fourth flow in one or more statistical time periods, and the throughput of the fourth flow in each statistical time period is obtained based on the transmission duration and traffic size of one or more message groups in the multiple message groups within the statistical time period at the end of transmission, and the transmission duration of each message group is obtained based on the end of transmission and the start of transmission of the message group.

[0078] Optionally, the measurement result includes flow rates of the one or more flows.

[0079] In a seventh aspect, a service flow analysis device is provided, which may be a network management device, including: a processor and a memory;

[0080] The memory is used to store a computer program, wherein the computer program includes program instructions;

[0081] The processor is used to call the computer program to implement the method in the above-mentioned first aspect and its various embodiments.

[0082] In an eighth aspect, a service traffic analysis device is provided, which may be a service platform and includes: a processor and a memory;

[0083] The memory is used to store a computer program, wherein the computer program includes program instructions;

[0084] The processor is used to call the computer program to implement the method in the above-mentioned second aspect and its various embodiments.

[0085] In a ninth aspect, a service flow analysis device is provided, which may be a network device, including: a processor and a memory;

[0086] The memory is used to store a computer program, wherein the computer program includes program instructions;

[0087] The processor is used to call the computer program to implement the method in the above-mentioned third aspect and its various embodiments.

[0088] In the tenth aspect, a business traffic analysis system is provided, comprising: a network management device and multiple network devices, wherein the network management device is used to execute the method in the above-mentioned first aspect and its various embodiments, and the network device is used to execute the method in the above-mentioned third aspect and its various embodiments.

[0089] Optionally, the system further includes a service platform, and the service platform is used to execute the method in the above-mentioned second aspect and its various embodiments.

[0090] In the eleventh aspect, a computer-readable storage medium is provided, on which instructions are stored. When the instructions are executed by the processor of the network management device, the method of the above-mentioned first aspect and its various embodiments is implemented; or, when the instructions are executed by the processor of the service platform, the method of the above-mentioned second aspect and its various embodiments is implemented; or, when the instructions are executed by the processor of the network device, the method of the above-mentioned third aspect and its various embodiments is implemented.

[0091] In the twelfth aspect, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the method in the above-mentioned first aspect and its various embodiments, or implements the method in the above-mentioned second aspect and its various embodiments, or implements the method in the above-mentioned third aspect and its various embodiments.

[0092] In a thirteenth aspect, a chip is provided. The chip includes a programmable logic circuit and / or program instructions. When the chip is running, it implements the method of the first aspect and its respective embodiments, or the method of the second aspect and its respective embodiments, or the method of the third aspect and its respective embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] FIG1 is a diagram of a distributed training task provided by an embodiment of the present application;

[0094] FIG2 is a schematic diagram of an implementation scenario provided by an embodiment of the present application;

[0095] FIG3 is a flow chart of a method for analyzing service traffic according to an embodiment of the present application;

[0096] FIG4 is a schematic diagram of a synchronization measurement result provided in an embodiment of the present application;

[0097] FIG5 is a schematic diagram of a throughput performance measurement result provided in an embodiment of the present application;

[0098] FIG6 is a flow chart of another service flow analysis method provided in an embodiment of the present application;

[0099] FIG7 is a schematic structural diagram of a service flow analysis device provided in an embodiment of the present application;

[0100] FIG8 is a schematic structural diagram of another service flow analysis device provided in an embodiment of the present application;

[0101] FIG9 is a schematic structural diagram of another service flow analysis device provided in an embodiment of the present application;

[0102] FIG10 is a schematic diagram of the hardware structure of a network device provided in an embodiment of the present application;

[0103] FIG11 is a schematic diagram of the hardware structure of a network management device / service platform provided in an embodiment of the present application. DETAILED DESCRIPTION

[0104] In order to make the objectives, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0105] Current networks often experience service interruptions and congestion due to network conditions, human intervention, or equipment defects. These traffic transmission issues directly impact service quality. Since most services (such as distributed training tasks, financial services, and banking services) are sensitive to quality issues, analyzing the transmission of service traffic within the network is essential. This allows for timely fault location and congestion locating after service traffic interruptions or congestion, allowing for appropriate isolation or recovery measures for the relevant network devices or transmission links.

[0106] Currently, network measurement primarily focuses on statistics related to network traffic, bandwidth, explicit congestion notification (ECN), and priority flow control (PFC). Most network measurements are performed at the flow granularity. For example, a flow can be identified using a binary, quadruple, or quintuple. Network management devices can capture the binary, quadruple, or quintuple of a passing packet from a network device to determine the flow to which the packet belongs. Combined with the network devices the flow's packets pass through, the flow's traffic volume, network path, link bandwidth, congestion, and whether PFC is enabled can be determined. A binary tuple includes the source and destination IP addresses. A quadruple includes the source and destination IP addresses, the source and destination ports, and the quintuple includes the source and destination IP addresses, the source and destination ports, and the transport layer protocol.

[0107] However, in some scenarios, a holistic analysis of service quality is required. Network measurement at the flow level is insufficient for comprehensive quality monitoring of services with multiple flows. For example, if a service involves multiple sources and / or destinations, the multiple flows of that service have multiple transmission paths within the network. Currently, there is no solution that allows simultaneous network measurement of multiple flows at the service level to analyze the overall network traffic for that service.

[0108] For example, distributed training tasks, as a distributed computing task, are a typical example of a service involving multiple business flows. With the development of artificial intelligence (AI) and big data technologies, the computing and data resources available for models are increasing, and the scale of model parameters and computational requirements are also increasing. This makes model training a very computationally intensive and time-consuming task. Distributed training has become a mainstream technology for improving model training efficiency. Distributed training tasks typically involve multiple computing nodes, each of which has a portion of data and model parameters. Multiple computing nodes need to exchange data over a network to jointly train the model. Currently, there are two basic approaches to distributed parallel model training: data parallelism (DP) and model parallelism (MP). Model parallelism approaches are categorized by how the model is partitioned: pipeline parallelism (PP) and tensor parallelism (TP). Data parallelism evenly distributes a batch of training data across multiple computing nodes, with each computing node maintaining a complete copy of the model. After each training session, computing nodes need to synchronize gradients to ensure that the model parameters stored on all computing nodes are identical. Model parallelism achieves the purpose of reducing memory usage during model training by splitting the model and placing it on multiple computing nodes. Pipeline parallelism splits the model by layer, and tensor parallelism adopts a complex intra-layer tensor splitting method. Among them, the data parallel (DP) stage and the pipeline parallel (PP) stage involve communication between different computing nodes (inter-machine communication), and the tensor parallel (TP) stage involves communication between different processing units within a single computing node (intra-machine communication). For example, Figure 1 is a distributed training task view provided by an embodiment of the present application. As shown in Figure 1, the distributed training task involves 4 computing nodes, which are respectively denoted as computing nodes N0 to N3. Each computing node Nx (x is any integer from 0 to 3) includes 4 graphics processing units (GPUs), denoted as GPU0 to GPU3. In the DP stage, each computing node Nx is allocated a portion of training data, and multiple computing nodes communicate gradients synchronously through the network. In the PP stage, multiple computing nodes exchange model data through the network. In the TP stage, multiple GPUs within the computing node exchange model data through bus technology, which can be, for example, peripheral component interconnect express (PCIe) technology or Nvlink technology.

[0109] In distributed training tasks, data exchange between computing nodes is typically implemented using collective communication. A distributed training task can include one or more collective communications. Collective communication refers to the coordinated communication between multiple processes in a distributed computing environment to complete certain specific tasks. Because a collective communication involves multiple interactive operations between computing nodes, all computing nodes participating in the collective communication must complete the interaction process for the collective communication to succeed and the distributed training task to continue. Otherwise, the distributed training task may slow down or become stuck. Therefore, it is necessary to perform network measurements at the granularity of distributed training tasks or collective communications.

[0110] The present application provides a technical solution for implementing network measurement with service as the granularity. In this technical solution, a network management device obtains the measurement task information of a service and issues a measurement command to multiple network devices in the network to instruct each network device that receives the measurement command to respectively count the traffic characteristics of one or more flows flowing through itself. The service includes multiple service flows, and the measurement task information of the service includes the device identifications of multiple terminal devices associated with the service, and the multiple terminal devices include the sending devices and receiving devices of the multiple service flows. Afterwards, after the network management device receives the measurement results reported by the multiple network devices, it determines the traffic analysis results corresponding to the service based on the measurement results sent by the multiple network devices and the device identifications of the multiple terminal devices. The measurement results sent by each network device respectively include the traffic characteristics corresponding to the flow identifications of one or more flows flowing through the network device. The flow identification of each flow includes the device identification of the sending device of the flow and the device identification of the receiving device of the flow. The present application collects the traffic characteristics corresponding to the flow identifiers of flows passing through multiple network devices in the network through a network management device, so that the network management device can determine the multiple business flows of a business with multiple business flows based on the collected flow identifiers and the device identifier in the measurement task information of the business, and then determine the traffic analysis results corresponding to the business based on the traffic characteristics of the multiple business flows, thereby realizing network measurement with business as the granularity, which is helpful for monitoring and analyzing business quality.

[0111] In the present application, a terminal device associated with a service refers to a terminal device that sends and / or receives one or more service flows of the service. Optionally, the service in the present application is a collective communication or a distributed training task. In the scenario where the service is a collective communication, the multiple terminal devices associated with the service refer to the multiple computing nodes participating in the collective communication, and the multiple computing nodes belong to a collective communication domain. Accordingly, the device identifier in the measurement task information is used to define all computing resources involved in the collective communication. Alternatively, in the scenario where the service is a distributed training task, the terminal device associated with the service refers to the computing node participating in the distributed training task, and the device identifier in the measurement task information is used to define all computing resources involved in the distributed training task. Optionally, a distributed training task generally includes multiple stages of collective communication, and the computing resources involved in the collective communication of different stages may be the same or different, that is, a distributed training task may correspond to one or more collective communication domains. The traffic analysis results in the present application may be presented at the granularity of collective communication. For example, for a distributed training task including multiple collective communications, the traffic analysis results corresponding to the distributed training task may include the traffic analysis results corresponding to the multiple collective communications.

[0112] The following is a detailed introduction to the technical solution of this application from multiple perspectives such as implementation scenarios, method flow, software devices, and hardware devices.

[0113] The following is an example of an implementation scenario of the embodiment of the present application.

[0114] For example, Figure 2 is a schematic diagram of an implementation scenario provided by an embodiment of the present application. As shown in Figure 2, the implementation scenario includes a network management device 201, multiple terminal devices 202A-202C (collectively referred to as terminal devices 202), and multiple network devices 203A-203E (collectively referred to as network devices 203) in a communication network. The number of terminal devices and network devices in Figure 2 is for illustrative purposes only and does not limit the implementation scenario of the embodiment of the present application.

[0115] The network management device 201 can be a server, a server cluster consisting of several servers, a cloud computing platform, or a network controller. The terminal device 202 can be a physical device such as a host or server, or a logical device such as a virtual machine obtained by virtualizing the computing resources of a computer device. The network device 203 can be a switch, a router, or a firewall. The network management device 201 and the network devices 203 are connected via a wired network or a wireless network. The network management device 201 is used to manage the network devices 203 in the communication network. For example, the network management device 201 can issue a measurement command to the network device 203 to instruct the network device 203 to count the traffic characteristics of one or more flows flowing through it, and receive and process the measurement results from the network device 203. Multiple terminal devices 202 communicate with each other through one or more network devices 203 in the communication network.

[0116] Optionally, when the implementation scenario shown in FIG2 is a distributed training task scenario or a collective communication scenario, the terminal device 202 is a computing node. The computing node may include one or more processors selected from a central processing unit (CPU), a GPU, a tensor processing unit (TPU), or a neural processing unit (NPU), and the computing node may be a server with computing capabilities.

[0117] The communication network provided in the embodiments of the present application may be a data center network (DCN), a metropolitan area network (MAN), a wide area network (WAN), a campus network, a virtual local area network (VLAN), or a virtual extensible local area network (VXLAN). The embodiments of the present application do not limit the type of communication network. For example, the communication network may be an RDMA over converged Ethernet (RoCE) network.

[0118] Optionally, the communication network provided in the embodiment of the present application can adopt a two-layer network architecture. The communication network includes a convergence layer and an access layer. The communication network can also be called a two-layer network. The convergence layer is the high-speed switching backbone of the communication network, and the access layer is used to connect terminal devices to the communication network. The network devices located in the access layer can be called access network devices, and the network devices located in the convergence layer can be called convergence network devices. For example, referring to Figure 2, network device 203A and network device 203B are located in the convergence layer and are convergence network devices. Network device 203C, network device 203D and network device 203E are located in the access layer and are access network devices. Each terminal device 202 is connected to network device 203C, network device 203D and network device 203E respectively. Network device 203C, network device 203D and network device 203E are connected to network device 203A and network device 203B respectively. A communication network adopting a two-layer network architecture can be, for example, a fat tree network, also known as a leaf-spine network.

[0119] Alternatively, the communication network provided in the embodiments of the present application may also adopt a three-layer network architecture. In a three-layer network architecture, the communication network includes a core layer, an aggregation layer, and an access layer. This communication network may also be referred to as a three-layer network. The core layer is the high-speed switching backbone of the communication network, the aggregation layer is used to provide aggregation connections (connecting the access layer and the core layer), and the access layer is used to connect terminal devices to the communication network.

[0120] Optionally, please continue to refer to Figure 2. The implementation scenario also includes a business platform 204. The business platform 204 can be a server, or a server cluster consisting of several servers, or a cloud computing platform. The business platform 204 is used to provide a human-computer interaction interface. Optionally, the business platform 204 includes a scheduling module, a tenant management module, and an operation and maintenance monitoring module. Among them, the scheduling module is used to allocate computing resources to the business. The tenant management module is used to manage tenant information. The operation and maintenance monitoring module is used to monitor the quality of the business. The business platform 204 is connected to the network management device 201 via a wired network or a wireless network. The business platform 204 can send business measurement tasks to the network management device 201, and receive and display the business measurement results sent by the network management device 201. Optionally, the business platform 204 can be an AI business operation and maintenance platform.

[0121] The following is an example of the method flow of the embodiment of the present application.

[0122] For example, Figure 3 is a flow chart of a service traffic analysis method provided in an embodiment of the present application. As shown in Figure 3, method 300 includes but is not limited to the following steps 301 to 304. Method 300 can be applied to the network management device 201 in the implementation scenario shown in Figure 2.

[0123] Step 301: The network management device obtains measurement task information of a service, where the measurement task information includes device identifiers of multiple terminal devices associated with the service. The service includes multiple service flows, and the multiple terminal devices associated with the service include sending devices and receiving devices of the multiple service flows.

[0124] Optionally, the device identification of the terminal device can be represented by the IP address of the terminal device. Alternatively, the device identification of the terminal device can also be represented by the IP address of the terminal device and the port identifier associated with the service on the terminal device. For example, a terminal device is associated with multiple services, and different services use different ports of the terminal device. Using the IP address and port identifier as the device identification can distinguish the resources provided by the same terminal device for different services. Alternatively, the terminal device is a virtual resource obtained by virtualizing the computing resources of a computer device. The device identification of the terminal device can be represented by a virtual resource identifier, such as the IP address assigned to the virtual resource. The embodiment of the present application does not limit the representation method of the device identification of the terminal device. The following embodiments take the device identification of the terminal device represented by the IP address of the terminal device as an example for explanation. Accordingly, the measurement task information includes the IP addresses of the multiple terminal devices associated with the service. These IP addresses can be the source IP addresses of the service flow of the service, or can be the destination IP addresses of the service flow of the service.

[0125] Optionally, the business flow of the above-mentioned business is an RDMA flow. RDMA is a message-based transmission protocol. In the embodiment of the present application, the source IP address and the destination IP address can be used to identify the business flow, and a business flow usually includes multiple message groups used to complete multiple data transmissions. When the amount of data transmitted in a single data transmission is large, it is usually necessary to split the data and transmit it by multiple messages, and the multiple messages are used to complete a message group for the data transmission. For RDMA flow, completing a data transmission means completing the transmission of a message, splitting a message into multiple messages for transmission, where the first message is the first message and the last message is the last message. The network device can identify a data transmission based on the received first message and last message.

[0126] Optionally, the measurement task information also includes a traffic analysis task, which is used to indicate the type of traffic analysis. Optionally, the traffic analysis task indicates analysis of one or more of the following: time synchronization of multiple service flows of the same service (time synchronization is used to reflect whether the flow completion time of the multiple service flows changes over time), the throughput performance of network devices in the network for multiple service flows of the same service, and the load balancing of multiple network devices in the network for the service. The traffic analysis task may also indicate analysis of the network paths of multiple service flows of the same service in the network, etc.

[0127] Optionally, the aforementioned service is a distributed training task or collective communication, where the distributed training task includes one or more collective communications. Because collective communication has the characteristics of large single-stream bandwidth, synchronous bursts, and communication efficiency dependent on the slowest stream, measuring the time synchronization of multiple streams in the collective communication, the throughput performance of network devices for each stream in the collective communication, and the load balancing of multiple network devices for the collective communication facilitates analysis of the communication quality of the collective communication.

[0128] Optionally, the network management device can obtain service measurement task information in various ways. For example, the network management device may receive a service measurement task sent by the service platform, where the service measurement task includes the service measurement task information. Alternatively, the service measurement task information may be manually input into the network management device. Alternatively, the network management device may proactively obtain the service measurement task information after a service failure occurs.

[0129] Step 302: The network management device sends a measurement command to multiple network devices in the network. The measurement command is used to instruct the network device that receives the measurement command to count traffic characteristics of one or more flows passing through the network device.

[0130] Optionally, the measurement command may indicate the specific type of traffic characteristics that the network device needs to count. For example, a flow includes multiple packet groups, each used to complete multiple data transmissions. The measurement command indicates that the traffic characteristics to be counted include the start and end times of each data transmission. Alternatively, the measurement command indicates that the traffic characteristics to be counted include the flow completion time. Alternatively, the measurement command indicates that the traffic characteristics to be counted include the start and end times of each data transmission, and the flow rate. Alternatively, the measurement command indicates that the traffic characteristics to be counted include throughput. Alternatively, the measurement command indicates that the traffic characteristics to be counted include flow rate.

[0131] Optionally, in the case where the measurement task information includes a traffic analysis task, the network management device may generate a measurement command based on the traffic analysis task, and the traffic characteristics indicated by the measurement command match the traffic analysis type indicated by the traffic analysis task. For example, the traffic analysis task indicates the time synchronization of multiple service flows of the analysis service, and accordingly, the measurement command indicates that the traffic characteristics to be counted include the start time and end time of each data transmission in the flow, or the flow completion time. For another example, the traffic analysis task indicates the throughput performance of multiple service flows of the analysis service by network devices in the network, and accordingly, the measurement command indicates that the traffic characteristics to be counted include the start time, end time and traffic size of each data transmission in the flow, or the throughput. For another example, the traffic analysis task indicates the load balancing of multiple network devices in the analysis network for the analysis service, and accordingly, the measurement command indicates that the traffic characteristics to be counted include the traffic size.

[0132] Alternatively, the network device may be pre-configured with a specific type of traffic feature to be counted, and the measurement command is used to trigger the network device to start performing traffic feature statistics without indicating the specific type of traffic feature to be counted by the network device.

[0133] Optionally, the measurement command further includes a measurement indication, where the measurement indication includes one or more of the following: a source IP address set, a destination IP address set, an inbound direction indication, an outbound direction indication, or an interface indication. Optionally, the device identifiers of the multiple terminal devices in the measurement task information are represented by IP addresses, and the source IP address set and the destination IP address set respectively include the IP addresses of one or more terminal devices among the multiple terminal devices.

[0134] The source IP address set is used to instruct the network device that receives the measurement command to calculate traffic characteristics of flows passing through the network device whose source IP addresses belong to the source IP address set. If the measurement command does not include the source IP address set, the network device does not filter flows passing through the network device based on source IP addresses, but instead calculates traffic characteristics of flows passing through the network device with any source IP address.

[0135] The destination IP address set instructs the network device that receives the measurement command to calculate traffic characteristics for flows passing through the network device whose destination IP addresses fall within the destination IP address set. If the measurement command does not include the destination IP address set, the network device does not filter flows passing through the network device based on their destination IP addresses. Instead, it calculates traffic characteristics for flows with any destination IP address passing through the network device.

[0136] The inbound direction indication is used to instruct the network device that has received the measurement command to collect statistics on traffic characteristics of one or more flows received by the network device.

[0137] The outgoing direction indication is used to instruct the network device that has received the measurement command to collect statistics on traffic characteristics of one or more flows sent by the network device.

[0138] If the measurement command includes only an inbound indication, the network device only counts the traffic characteristics of one or more flows received by the network device. If the measurement command includes only an outbound indication, the network device only counts the traffic characteristics of one or more flows sent by the network device. If the measurement command includes neither an inbound indication nor an outbound indication, or if the measurement command includes both an inbound indication and an outbound indication, the network device counts the traffic characteristics of one or more flows received by the network device and the traffic characteristics of one or more flows sent by the network device.

[0139] The interface indication is used to instruct the network device that receives the measurement command to count the traffic characteristics of flows flowing through one or more specified interfaces of the network device. If the measurement command does not include the interface indication, the network device counts the traffic characteristics of flows flowing through all its interfaces.

[0140] In an embodiment of the present application, the network management device carries a measurement indication in a measurement command, so that the network device can perform targeted statistics on the traffic characteristics of some flows, reduce the network device's statistics on the traffic characteristics of useless flows, and thus save the processing resources of the network device.

[0141] Optionally, the measurement command may also include a validity period for the measurement task, such as permanent validity or validity for a period of time. If the measurement command indicates permanent validity, the network device will continue to collect traffic statistics on flows passing through it until it receives a measurement termination command from the network management device. If the measurement command indicates validity for a period of time, the network device will only collect traffic statistics on flows passing through it during the validity period. After the validity period expires, the network device will automatically cease executing the traffic statistics task.

[0142] Step 303: The network management device receives the measurement results sent by the multiple network devices. The measurement results sent by each network device include the traffic characteristics corresponding to the flow identifiers of one or more flows flowing through the network device. The flow identifier of each flow includes the device identifier of the sending device of the flow and the device identifier of the receiving device of the flow.

[0143] Optionally, after the network device receives the measurement command sent by the network management device, it counts the traffic characteristics of one or more flows flowing through itself according to the measurement command, and then sends the measurement results corresponding to the measurement command to the network management device. Optionally, if the flow flowing through the network device is an RDMA flow, the flow identifier can also include a queue pair (QP) number. For example, the flow identifier can be represented by the source IP address of the flow, the destination IP address of the flow, and the QP number. When transmitting messages between terminal devices based on the RDMA protocol, the communicating parties first need to establish a QP, where the sender establishes a send queue (SQ) and the receiver establishes a receive queue (RQ), and the communicating parties communicate based on the QP. Since a terminal device may be associated with multiple services, when the IP address of the terminal device is used as the device identifier of the terminal device, if two services have service flows with the same source IP address and destination IP address, the network management device cannot distinguish the service flows of different services by the source IP address and destination IP address. Terminal devices need to establish QPs separately for different services to transmit the messages of each service. The QP numbers of different services are usually different. Therefore, the network device adds the QP number to the flow identifier reported to the network management device, so that the network management device can distinguish the service flows of different services with the same source IP address and destination IP address according to the flow identifier.

[0144] Optionally, when the measurement command sent by the network management device to the network device in step 302 includes an interface indication, the measurement result sent by the network device to the network management device may further include an interface identifier, which is used to indicate the interface through which the corresponding flow passes on the network device.

[0145] Optionally, the measurement result sent by the network device to the network management device also includes a device identifier of the network device. The device identifier of the network device may be, for example, the IP address of the network device, the Media Access Control (MAC) address of the network device, or the hardware address of the network device, which is information that uniquely identifies the network device.

[0146] Step 304: The network management device determines a traffic analysis result corresponding to the service based on the measurement results sent by the multiple network devices and the device identifications of the multiple terminal devices.

[0147] Optionally, step 304 may be implemented by, for each measurement result sent by a network device, the network management device determining, from one or more flows passing through the network device, a service flow belonging to the service, based on the flow identifier in the measurement result and the device identifiers of multiple terminal devices associated with the service in the measurement task information. The network management device determines a traffic analysis result corresponding to the service based on the traffic characteristics of the multiple service flows in the measurement results sent by the multiple network devices. The network management device may determine, as a service flow for the service, a flow whose sending and receiving devices both belong to the multiple terminal devices.

[0148] In an embodiment of the present application, the network management device collects traffic characteristics corresponding to the flow identifiers of flows passing through multiple network devices in the network, so that the network management device can determine the multiple business flows of the business based on the collected flow identifiers and the device identifier in the measurement task information of the business for a business with multiple business flows, and then determine the traffic analysis results corresponding to the business based on the traffic characteristics of the multiple business flows, thereby realizing network measurement with business as the granularity, which is helpful for monitoring and analyzing business quality.

[0149] Optionally, when the measurement task information includes a traffic analysis task, the network management device determines the traffic analysis result of the service under the traffic analysis type indicated by the traffic analysis task based on the measurement results sent by the multiple network devices and the device identifications of the multiple terminal devices associated with the service in the measurement task information.

[0150] The following embodiments of the present application provide examples of traffic characteristics counted by network devices and traffic analysis results determined by network management devices under different traffic analysis tasks.

[0151] In the first possible scenario, the traffic analysis task indicates analyzing the time synchronization of multiple service flows of a service. Accordingly, the traffic analysis result includes a synchronization measurement result, which includes the flow completion time of the multiple service flows of the service in multiple statistical time periods.

[0152] Optionally, the multiple statistical periods are sequentially continuous, i.e., the end time of the previous statistical period is the start time of the next statistical period. Alternatively, the multiple statistical periods are sequentially discontinuous, i.e., the start time of the next statistical period is after the end time of the previous statistical period.

[0153] In an embodiment of the present application, by measuring the flow completion time of multiple business flows of the same business in multiple statistical time periods, the changes in the flow completion time of multiple business flows over time are presented at the business granularity. The changes in the flow completion time of a single business flow over time can reflect the changes in the transmission quality of the business flow itself, and the changes in the flow completion time of multiple business flows over time can reflect the changes in the overall transmission quality of the business. This helps operation and maintenance personnel analyze whether the overall transmission quality of the business has deteriorated and the specific business flows that have caused the deterioration of the business transmission quality. Taking collective communication as an example, since collective communication has the characteristics of synchronous bursts and the communication efficiency depends on the slowest flow, by counting the changes in the flow completion time of all business flows in a collective communication over time, on the one hand, it can be determined whether the communication quality of the collective communication is stable, and on the other hand, it can be determined whether the flow completion time of a certain business flow is too long, resulting in low efficiency of the collective communication, thereby realizing the analysis of the communication quality of the collective communication at different granularities.

[0154] Optionally, for any of the multiple service flows of a service, if the difference between the flow completion time of the service flow in the third statistical period and the flow completion time of the service flow in the fourth statistical period is greater than a first threshold, the network management device determines that the transmission quality of the service flow has deteriorated. The third statistical period is located after the fourth statistical period in terms of timing. That is to say, if the flow completion time of a certain service flow becomes longer, it means that the transmission quality of the service flow has deteriorated. Accordingly, the synchronization measurement result may also include an indication of the service flow whose transmission quality has deteriorated among the multiple service flows of the service. The indication of a service flow may include the device identification of the sending device of the service flow (such as the source IP address of the service flow) and the device identification of the receiving device (such as the destination IP address of the service flow). The indication of a service flow may also include the transmission path of the service flow in the network, etc., so that the network location that causes the transmission quality of the service flow to deteriorate can be checked on the transmission path of the service flow, which helps to improve the analysis efficiency.

[0155] In a first implementation of the first possible scenario described above, the measurement command sent by the network management device to multiple network devices may indicate that the traffic characteristics to be counted include the start and end times of each data transmission in a flow, or pre-configure the network devices to count the start and end times of each data transmission in a flow passing through them. Accordingly, the measurement results sent by the network devices to the network management device include the start and end times of each data transmission in one or more flows passing through them. For example, the multiple network devices receiving the measurement command include a first network device, where the first network device is any one of the multiple network devices. The measurement results sent by the first network device to the network management device include first traffic characteristics of a first service flow. The first service flow is any one of multiple service flows of a service, and the first service flow includes multiple packet groups, each used to complete multiple data transmissions. The first traffic characteristics include the transmission start and end times of each of the multiple packet groups. In this implementation, the network management device may determine the flow completion time of the first service flow within the first statistical period based on the transmission duration of one or more packet groups within the multiple packet groups whose transmission end times fall within the first statistical period. The transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group. The transmission duration of a message group, that is, the completion time of a data transmission, can be equal to the value of the transmission end time of the message group minus the transmission start time. The first statistical period is any statistical period among multiple statistical periods. The method for determining the flow completion time of the first business flow in other statistical periods and the flow completion time of other business flows in multiple statistical periods can refer to the method for determining the flow completion time of the first business flow in the first statistical period, and the embodiments of the present application will not be repeated here.

[0156] Optionally, the flow completion time of a service flow within a statistical period is a statistical value of the transmission duration of all message groups in the service flow whose transmission ends within the statistical period, and the statistical value includes but is not limited to an average value, a median value, a maximum value, or a percentile value. For example, the network management device determines the flow completion time of the first service flow within the first statistical period based on the transmission duration of one or more message groups in the multiple message groups whose transmission ends within the first statistical period. The network management device may use the statistical value of the transmission duration of all message groups in the multiple message groups of the first service flow whose transmission ends within the first statistical period as the flow completion time of the first service flow within the first statistical period.

[0157] Under this implementation method, the network device only needs to count the start and end times of each data transmission in the flow passing through itself. The network management device performs traffic analysis on the service based on the traffic characteristics reported by the network device to determine the flow completion time of multiple business flows of the service in multiple statistical time periods. This implementation method has low requirements on the processing performance of the network device.

[0158] Optionally, the network device can periodically send measurement results to the network management device during the measurement task's validity period. The measurement results include the start and end times of data transmissions completed by one or more flows passing through the network device during the current period. For example, flow 1 passing through network device 1 completed three data transmissions during the current period. These three data transmissions correspond to three packet groups: packet group 11, packet group 12, and packet group 13. The transmission of packet group 11 started at time T11 and ended at time T12. The transmission of packet group 12 started at time T13 and ended at time T14. The transmission of packet group 13 started at time T15 and ended at time T16. Flow 2 passing through network device 1 completed two data transmissions during the current period. These two data transmissions correspond to two packet groups: packet group 21 and packet group 22. The transmission of packet group 21 started at time T21 and ended at time T22. The transmission of message group 22 begins at time T23 and ends at time T24. The source IP address of flow 1 is IP1, and the destination IP address is IP2. The source IP address of flow 2 is IP3, and the destination IP address is IP4. The measurement results sent by network device 1 to the network management device can be shown in Table 1.

[0159] Table 1

[0160] Referring to Table 1, assuming that IP1 and IP2 are the IP addresses of terminal devices associated with the service, that is, flow 1 is the service flow of the service to be analyzed, and T12, T14, and T16 are within the same statistical period, the network management device first calculates the transmission duration t11 = T12 - T11 of message group 11, the transmission duration t12 = T14 - T13 of message group 12, and the transmission duration t13 = T16 - T15 of message group 13. Then, the flow completion time of flow 1 within the statistical period is calculated as the average, median, maximum, or percentile value of t11, t12, and t13.

[0161] Alternatively, in the first implementation method of the first possible situation mentioned above, the traffic characteristics reported by the network device to the network management device may include, in addition to the start time and end time of each data transmission in the flow, or may include the start time and transmission duration of each data transmission in the flow, or may include the end time and transmission duration of each data transmission in the flow, or may include the start time, end time and transmission duration of each data transmission in the flow.

[0162] In a second implementation of the first possible scenario described above, the measurement command sent by the network management device to the network device may indicate that the traffic characteristics to be counted include flow completion time, or the network device may be pre-configured to count the flow completion time of flows passing through it within multiple statistical time periods. Accordingly, the measurement results sent by the network device to the network management device include the flow completion time of one or more flows passing through it. For example, the multiple network devices receiving the measurement command include a second network device, where the second network device is any one of the multiple network devices. The measurement results sent by the second network device to the network management device include the second traffic characteristics of the second service flow. The second service flow is any one of the multiple service flows of a service, and the second service flow includes multiple packet groups, each used to complete multiple data transmissions. The second traffic characteristics include the flow completion time of the second service flow within the second statistical time period. The flow completion time of the second service flow within the second statistical time period is derived based on the transmission duration of one or more packet groups within the multiple packet groups whose transmission end times are within the second statistical time period. The transmission duration of each packet group is derived based on the transmission end time and transmission start time of the packet group. The second statistical time period is any one of the multiple statistical time periods. The way in which a network device determines the flow completion time of a flow passing through itself within a statistical period can refer to the way in which a network management device determines the flow completion time of a service flow within a statistical period in the first implementation method of the first possible case mentioned above. The embodiments of this application will not be repeated here.

[0163] Optionally, the flow completion time of the second service flow in the second statistical period is a statistical value of the transmission duration of all message groups of the multiple message groups of the second service flow whose transmission end moments are within the second statistical period.

[0164] Under this implementation method, the network device counts the start and end times of each data transmission in the flow passing through itself, and performs traffic analysis to determine the flow completion time of the flow passing through itself within multiple statistical time periods. The network management device summarizes and counts the traffic characteristics reported by multiple network devices, and can obtain the flow completion time of multiple business flows of the business within multiple statistical time periods. This implementation method has low processing performance requirements for the network management device and can improve the efficiency of the network management device in analyzing the traffic of the business.

[0165] Optionally, the network device can periodically count the flow completion times of one or more flows passing through it within the effective period of the measurement task and report these to the network management device. The multiple statistical periods described above constitute multiple statistical cycles. The single measurement result sent by the network device to the network management device may include the flow completion times of the flows passing through it within one or more statistical periods. For example, network device 2 calculates that: the flow completion time of flow 3 passing through network device 2 within statistical period A1 is FCT1, and the flow completion time within statistical period A2 is FCT2; and the flow completion time of flow 4 passing through network device 2 within statistical period A1 is FCT3, and the flow completion time within statistical period A2 is FCT4. The source IP address of flow 3 is IP5, and the destination IP address is IP6. The source IP address of flow 4 is IP7, and the destination IP address is IP8. The single measurement result sent by network device 2 to the network management device may be as shown in Table 2.

[0166] Table 2

[0167] Optionally, each statistical period in Table 2 may be represented by the start time and end time of the statistical period. If multiple statistical periods are consecutive in time sequence, each statistical period in Table 2 may also be represented by the start time of the statistical period, with the start time of the next statistical period being the end time of the current statistical period, or each statistical period in Table 2 may also be represented by the end time of the statistical period, with the start time of the current statistical period being the end time of the previous statistical period.

[0168] Optionally, in a second implementation of the first possible situation, the measurement result sent by the network device to the network management device may further include the start time and end time of each data transmission in the flow passing through the network device.

[0169] Optionally, the synchronization measurement result may further include two or all of the start time, end time, and transmission duration of each data transmission in the multiple service flows of the service.

[0170] Optionally, the synchronization measurement result may also include the earliest flow start time of multiple business flows of the business and the latest flow end time of the multiple business flows. The flow start time of a business flow is the start time of the first data transmission in the business flow, and the flow end time of a business flow is the end time of the last data transmission in the business flow. For example, taking the above-mentioned multiple business flows belonging to a collective communication as an example, the earliest flow start time of the multiple business flows is also the start time of the collective communication, and the latest flow end time of the multiple business flows is also the end time of the collective communication.

[0171] Taking the above-mentioned service as a collective communication as an example, assuming that the IP addresses of the computing nodes participating in the collective communication include IP1 to IP4, the network management device determines that the collective communication includes 8 service flows based on the measurement results reported by multiple network devices. The communication pairs involved in the 8 service flows include IP1→IP2 (the source IP address of the service flow is IP1, and the destination IP address is IP2), IP2→IP1 (the source IP address of the service flow is IP2, and the destination IP address is IP1), IP3→IP4 (the source IP address of the service flow is IP3, and the destination IP address is IP4), IP4→IP3 (the source IP address of the service flow is IP4, and the destination IP address is IP3), IP1→IP3 (the source IP address of the service flow is IP1, and the destination IP address is IP3), IP3→IP1 (the source IP address of the service flow is IP3, and the destination IP address is IP1), IP2→IP4 (the source IP address of the service flow is IP2, and the destination IP address is IP4), and IP4→IP2 (the source IP address of the service flow is IP4, and the destination IP address is IP2). Taking the synchronization measurement results as an example, including the flow completion time of multiple business flows of collective communication in multiple statistical time periods, the start time and end time of collective communication, and the start time, end time and duration of each data transmission in the business flow of collective communication, the synchronization measurement results can be shown in Table 3.

[0172] Table 3

[0173] Alternatively, the synchronization measurement result can also be expressed in the form of a graph. For example, Figure 4 is a schematic diagram of a synchronization measurement result provided by an embodiment of the present application. The horizontal axis represents time (t) and the vertical axis represents flow completion time (FCT). As shown in Figure 4, within statistical time periods A5 and A6, the flow completion time of communication pairs IP1→IP3 and IP3→IP1 is significantly longer. At this time, it can be considered that the transmission quality of the service flow with source IP address IP1 and destination IP address IP3 and the service flow with source IP address IP3 and destination IP address IP1 has deteriorated, and the root cause of the deterioration of the transmission quality of the service flow can be further located in the network, thereby improving the stability and reliability of the collective communication.

[0174] In the second possible scenario, the traffic analysis task instructs the user to analyze the throughput performance of multiple service flows of a network device in the network. Accordingly, the traffic analysis results include throughput performance measurement results, which include the throughput of each service flow passing through each network device in the network over multiple statistical time periods.

[0175] In the embodiment of the present application, the flow throughput performance mainly reflects the size of the traffic flow received or sent by the network device. By measuring the throughput of multiple business flows of the same business by the network device in multiple statistical time periods, the throughput of the network device to the business flow changes over time with the business as the granularity, wherein the change of the throughput of the network device to a single business flow over time can reflect the change of the transmission quality of the business flow itself, and the change of the throughput of multiple network devices to multiple business flows over time can reflect the change of the overall transmission quality of the business, which helps the operation and maintenance personnel to analyze whether the overall transmission quality of the business has deteriorated and the specific business flow that causes the deterioration of the business transmission quality. Taking collective communication as an example, since collective communication has the characteristics of synchronous burst and the communication efficiency depends on the slowest flow, by counting the changes of the throughput of the network device to all business flows in a collective communication over time, on the one hand, it can be judged whether the communication quality of the collective communication is stable, and on the other hand, it can be judged whether the efficiency of the collective communication is low due to the small throughput of a certain network device to the business flow, thereby realizing the analysis of the communication quality of the collective communication at different granularities.

[0176] Optionally, for any of the multiple service flows of a service, if the difference between the throughput of the service flow during the seventh statistical period and the throughput of the service flow during the eighth statistical period by any network device through which the service flow passes is greater than a second threshold, the network management device determines that the transmission quality of the service flow has deteriorated. The eighth statistical period is located after the seventh statistical period in terms of time sequence. That is, if the throughput of the network device for a certain service flow decreases, it indicates that the transmission quality of the service flow has deteriorated. Accordingly, the throughput performance measurement result may also include an indication of the service flow among the multiple service flows whose transmission quality has deteriorated. The indication of a service flow may include the device identifier of the sending device of the service flow (such as the source IP address of the service flow) and the device identifier of the receiving device (such as the destination IP address of the service flow). The indication of a service flow may also include the transmission path of the service flow in the network and the device identifier of the network device that counts the throughput of the service flow, etc. In this way, the network location that causes the transmission quality of the service flow to deteriorate can be identified on the transmission path of the service flow, which helps to improve analysis efficiency.

[0177] In the first implementation of the second possible scenario described above, the measurement command sent by the network management device to multiple network devices may indicate that the traffic characteristics to be counted include the start time, end time, and traffic volume of each data transmission in the flow, or pre-configure the network device to count the start time, end time, and traffic volume of each data transmission in the flow passing through it. Accordingly, the measurement results sent by the network device to the network management device include the start time, end time, and traffic volume of each data transmission in one or more flows passing through it. For example, the multiple network devices receiving the measurement command include a third network device, where the third network device is any one of the multiple network devices. The measurement results sent by the third network device to the network management device include third traffic characteristics of a third service flow. The third service flow is any one of the multiple service flows of a service, and the third service flow includes multiple packet groups used to complete multiple data transmissions. The third traffic characteristics include the transmission start time, transmission end time, and traffic volume of each of the multiple packet groups. Under this implementation, the network management device can determine the throughput of the third service flow of the third network device in the fifth statistical period based on the transmission duration and flow size of one or more message groups among the multiple message groups whose transmission end moments are within the fifth statistical period. The transmission duration of each message group is obtained based on the transmission end moment and transmission start moment of the message group. The fifth statistical period is any statistical period among the multiple statistical periods. The method for determining the throughput of the third service flow of the third network device in other statistical periods and the throughput of the service flow flowing through the other network devices in multiple statistical periods can all refer to the method for determining the flow completion time of the third service flow of the third network device in the fifth statistical period, and the embodiments of the present application will not be repeated here.

[0178] Optionally, the throughput of a service flow of a network device in a statistical period may be a ratio of a total flow size of the service flow transmitted by the network device in the statistical period to a flow completion time of the service flow in the statistical period.

[0179] Under this implementation method, the network device only needs to count the start time, end time and flow size of each data transmission in the flow passing through itself. The network management device performs traffic analysis on the service based on the traffic characteristics reported by multiple network devices to determine the throughput of multiple service flows of the service by multiple network devices in multiple statistical time periods. This implementation method has low requirements on the processing performance of the network equipment.

[0180] Optionally, within the effective period of the measurement task, the network device can periodically send measurement results to the network management device. These measurement results include the start time, end time, and flow rate of data transmissions completed by one or more flows passing through the network device during the current period. For example, flow 5 passing through network device 3 completed three data transmissions during the current period. These three data transmissions correspond to three packet groups: packet group 51, packet group 52, and packet group 53. The transmission of packet group 51 started at time T51 and ended at time T52, with a flow rate of L1. The transmission of packet group 52 started at time T53 and ended at time T54, with a flow rate of L2. The transmission of packet group 53 started at time T55 and ended at time T56, with a flow rate of L3. Flow 6 passing through network device 3 completed two data transmissions during the current period. These two data transmissions correspond to two packet groups: packet group 61 and packet group 62. The transmission of message group 61 starts at T61 and ends at T62, with a flow rate of L4. The transmission of message group 62 starts at T63 and ends at T64, with a flow rate of L5. The source IP address of flow 5 is IP1, and the destination IP address is IP2. The source IP address of flow 6 is IP3, and the destination IP address is IP4. The measurement results sent by network device 3 to the network management device can be shown in Table 4.

[0181] Table 4

[0182] Referring to Table 4, assuming that IP1 and IP2 are the IP addresses of the terminal devices associated with the service, that is, flow 5 is the service flow of the service to be analyzed, and T52, T54, and T56 are within the same statistical period, the network management device first calculates the transmission duration t51 of message group 51 = T52 - T51, the transmission duration t52 of message group 52 = T54 - T53, and the transmission duration t53 of message group 53 = T56 - T55. Then, the flow completion time t5 of flow 5 within the statistical period is calculated as the average, median, or maximum value of t51, t52, and t53. Finally, the throughput of network device 3 for flow 5 within the statistical period is calculated to be (L1+L2+L3) / t5. Alternatively, other calculation methods may be used. For example, the throughput corresponding to time T52 is TH52 = L1 / t51, the throughput corresponding to time T54 is TH54 = L2 / t52, and the throughput corresponding to time T56 is TH56 = L3 / t53. The throughput of flow 5 of network device 3 during the statistical period is the average of TH52, TH54, and TH56.

[0183] Alternatively, in the first implementation method of the above-mentioned second possible situation, the traffic characteristics reported by the network device to the network management device may include, in addition to the start time, end time and traffic size of each data transmission in the flow, or the start time, transmission duration and traffic size of each data transmission in the flow, or the end time, transmission duration and traffic size of each data transmission in the flow, or the start time, end time, transmission duration and traffic size of each data transmission in the flow, or the end time and throughput of each data transmission.

[0184] In a second implementation of the second possible scenario described above, the network management device may send a measurement command to multiple network devices to indicate that the traffic characteristics to be counted include throughput, or pre-configure the network devices to count the throughput of flows passing through them over multiple statistical time periods. Accordingly, the measurement results sent by the network devices to the network management device include the throughput of one or more flows passing through them. For example, the multiple network devices receiving the measurement command include a fourth network device, where the fourth network device is any one of the multiple network devices. The measurement results sent by the fourth network device to the network management device include a fourth traffic characteristic of a fourth service flow. The fourth service flow is any one of multiple service flows, and the fourth service flow includes multiple packet groups, each used to complete multiple data transmissions. The fourth traffic characteristic includes the throughput of the fourth service flow by the fourth network device over a sixth statistical time period. The throughput of the fourth service flow over the sixth statistical time period is determined based on the transmission duration and traffic volume of one or more packet groups among the multiple packet groups whose transmission end times fall within the sixth statistical time period. The transmission duration of each packet group is determined based on the transmission end time and transmission start time of the packet group. The sixth statistical time period is any one of the multiple statistical time periods. The way in which a network device determines the throughput of a flow passing through itself during a statistical period can refer to the way in which a network management device determines the throughput of a service flow during a statistical period in the first implementation method of the second possible case mentioned above. The embodiments of this application will not be repeated here.

[0185] Under this implementation method, the network device counts the start time, end time and flow size of each data transmission in the flow passing through itself, and performs traffic analysis to determine the throughput of the flow passing through itself in multiple statistical time periods. The network management device summarizes and counts the traffic characteristics reported by multiple network devices, and can obtain the throughput of multiple business flows of multiple network devices for the business in multiple statistical time periods. This implementation method has low processing performance requirements for the network management device and can improve the efficiency of the network management device's traffic analysis of the business.

[0186] Optionally, the network device may periodically count the throughput of one or more flows passing through it and report it to the network management device within the effective period of the measurement task. The above-mentioned multiple statistical periods are multiple statistical cycles. The measurement results sent by the network device to the network management device may include the throughput of the flows passing through it within one or more statistical periods. For example, network device 4 calculates that: the throughput of flow 7 of network device 4 within statistical period A1 is F1, and the throughput within statistical period A2 is F2; ​​and the throughput of flow 8 of network device 4 within statistical period A1 is F3, and the throughput within statistical period A2 is F4. Among them, the source IP address of flow 7 is IP5, and the destination IP address is IP6. The source IP address of flow 8 is IP7, and the destination IP address is IP8. Then the measurement results sent by network device 4 to the network management device may be as shown in Table 5.

[0187] Table 5

[0188] Optionally, each statistical period in Table 5 may be represented by the start time and end time of the statistical period. If multiple statistical periods are consecutive in time sequence, each statistical period in Table 5 may also be represented by the start time of the statistical period, with the start time of the next statistical period being the end time of the current statistical period, or each statistical period in Table 5 may also be represented by the end time of the statistical period, with the end time of the current statistical period being the start time of the next statistical period.

[0189] Optionally, in the second possible scenario described above, the measurement command may further include an inbound direction indication, so that the network device only counts the start time, end time, and flow rate of each data transmission in the received flow in the inbound direction. In this way, the throughput of the network device for the service flow actually reflects the flow rate of the service flow received by the network device. Of course, the embodiments of the present application do not preclude the implementation of a method for calculating throughput by counting the start time, end time, and flow rate of each data transmission in the outbound flow in the outbound direction.

[0190] Optionally, the throughput performance measurement result also includes a traffic direction and / or an interface identifier of the network device. The traffic direction indicates whether the traffic flow being counted is a traffic flow received by the network device or a traffic flow sent by the network device. The interface identifier indicates the interface on the network device through which the traffic flow passes.

[0191] Taking the above-mentioned service as a collective communication as an example, assuming that the IP addresses of the computing nodes participating in the collective communication include IP1 to IP4, the network management device determines that the collective communication includes 8 service flows based on the measurement results reported by multiple network devices. The communication pairs involved in the 8 service flows include IP1→IP2 (the source IP address of the service flow is IP1, and the destination IP address is IP2), IP2→IP1 (the source IP address of the service flow is IP2, and the destination IP address is IP1), IP3→IP4 (the source IP address of the service flow is IP3, and the destination IP address is IP4), IP4→IP3 (the source IP address of the service flow is IP4, and the destination IP address is IP3), IP1→IP3 (the source IP address of the service flow is IP1, and the destination IP address is IP3), IP3→IP1 (the source IP address of the service flow is IP3, and the destination IP address is IP1), IP2→IP4 (the source IP address of the service flow is IP2, and the destination IP address is IP4), and IP4→IP2 (the source IP address of the service flow is IP4, and the destination IP address is IP2). Taking the throughput performance measurement results as an example, which include the throughput of the service flows passing through multiple network devices in the network in multiple statistical time periods, the interface identifiers of the interfaces through which the service flows pass on the network devices, and the traffic direction, the throughput performance measurement results can be shown in Table 6.

[0192] Table 6

[0193] As shown in Table 6, network device 1 is used to collect traffic statistics for service flows with a source IP address of IP1 and a destination IP address of IP2, as well as service flows with a source IP address of IP3 and a destination IP address of IP4. The inbound interface of the service flows with a source IP address of IP1 and a destination IP address of IP2 on network device 1 is interface 11, and the inbound interface of the service flows with a source IP address of IP3 and a destination IP address of IP4 on network device 1 is interface 12. Network device 2 is used to collect traffic statistics for service flows with a source IP address of IP2 and a destination IP address of IP1, as well as service flows with a source IP address of IP4 and a destination IP address of IP3. The inbound interface of the service flows with a source IP address of IP2 and a destination IP address of IP1 on network device 2 is interface 21, and the inbound interface of the service flows with a source IP address of IP4 and a destination IP address of IP3 on network device 2 is interface 22. Network device 3 is used to collect traffic statistics for service flows with a source IP address of IP1 and a destination IP address of IP3, as well as service flows with a source IP address of IP2 and a destination IP address of IP4. The inbound interface of the service flows with a source IP address of IP1 and a destination IP address of IP3 on network device 3 is interface 31, and the inbound interface of the service flows with a source IP address of IP2 and a destination IP address of IP4 on network device 3 is interface 32. Network device 4 is used to collect traffic statistics for service flows with a source IP address of IP3 and a destination IP address of IP1, as well as service flows with a source IP address of IP4 and a destination IP address of IP2. The inbound interface of the service flows with a source IP address of IP3 and a destination IP address of IP1 on network device 4 is interface 41, and the inbound interface of the service flows with a source IP address of IP4 and a destination IP address of IP2 on network device 4 is interface 42.

[0194] Alternatively, the throughput performance measurement results can also be represented in the form of a graph. For example, Figure 5 is a schematic diagram of a throughput performance measurement result provided by an embodiment of the present application. The horizontal axis represents time (t) and the vertical axis represents throughput. As shown in Figure 5, within statistical time periods A5 and A6, the throughput of communication pairs IP1→IP3 and IP3→IP1 is significantly reduced. At this time, it can be considered that the transmission quality of the service flow with a source IP address of IP1 and a destination IP address of IP3 and the service flow with a source IP address of IP3 and a destination IP address of IP1 has deteriorated, and the root cause of the deterioration of the transmission quality of the service flow can be further located in the network, thereby improving the stability and reliability of the collective communication.

[0195] Optionally, in the first and second possible scenarios, step 302 may be implemented by: the network management device sending measurement commands to multiple access network devices in the network, each of which is used to connect one or more terminal devices associated with the service to the network. For example, in the application scenario shown in FIG2 , the terminal devices associated with the service include terminal device 202A, terminal device 202B, and terminal device 202C, i.e., the measurement task information includes the device identifier of terminal device 202A, the device identifier of terminal device 202B, and the device identifier of terminal device 202C. In this case, the network management device may send measurement commands only to network device 203C, network device 203D, and network device 203E, respectively.

[0196] In the embodiments of the present application, since the service flows sent or received by the terminal devices associated with the service inevitably pass through the access network device, and the flow completion time and throughput of a service flow on different network devices when it is transmitted in the network are basically the same, when the traffic characteristics to be counted include flow completion time and / or throughput, it is sufficient to count the traffic characteristics of the flows passing through the access network device. In this way, while ensuring that the traffic characteristics of all service flows of the service are counted, the number of network devices involved in service traffic analysis can be reduced, saving communication resources between network devices and network management devices, as well as their respective processing resources.

[0197] In a third possible scenario, the traffic analysis task instructs analysis of the load balancing of services across multiple network devices in the network. Accordingly, the traffic analysis results include load balancing measurement results, which include the load balancing degree of the services across multiple network devices in the network. In this possible scenario, the measurement command sent by the network management device to the multiple network devices may indicate that the traffic characteristics to be counted include traffic volume, or the network devices may be pre-configured to count the traffic volume of flows passing through them. For each network device, the network management device may determine the cumulative traffic volume of the service on that network device based on the traffic volumes of all service flows in the measurement results sent by the network device. The cumulative traffic volume of a service on a network device is the sum of the traffic volumes of all service flows passing through the network device. The network management device determines the load balancing degree of the service on each of the multiple network devices based on the cumulative traffic volumes of the service. The load balancing degree is negatively correlated with the difference between the cumulative traffic volumes of the services across the multiple network devices. That is, the smaller the difference in the cumulative traffic volumes of the services across the multiple network devices, the higher the degree of load balancing.

[0198] Optionally, there may be multiple criteria for measuring the degree of load balancing. For example, if the ideal traffic load condition is that the cumulative traffic volume of the access network devices for the service is the same, then the load balancing degree may be determined based on the cumulative traffic volume of the service for multiple access network devices. Alternatively, if the ideal traffic load condition is that the cumulative traffic volume of the service for all network devices involved in the service is the same, then the load balancing degree may be determined based on the cumulative traffic volume of the service for all network devices.

[0199] Optionally, the network device may periodically send a measurement result to the network management device during the effective period of the measurement task. The measurement result includes the flow volume of one or more flows passing through the network device during the current period.

[0200] In the embodiments of the present application, load balancing primarily reflects the differences in traffic volumes for the same service across different network devices. A higher load balancing indicates a more even distribution of traffic across network devices when multiple service flows for that service are transmitted across the network, and a lower probability of network device congestion. By analyzing the load balancing of different network devices along the network path through which service flows pass, network management devices can help operations and maintenance teams analyze and adjust the transmission paths of service flows within a service, thereby improving load balancing across multiple network devices for the service.

[0201] Optionally, the load balancing measurement result may further include the cumulative traffic size of each of the multiple network devices for the service. For example, the load balancing measurement result may be as shown in Table 7.

[0202] Table 7

[0203] Alternatively, in the third possible scenario, step 302 may be implemented by the network management device sending measurement commands only to the multiple network devices along the network paths of the multiple service flows in the network. This reduces the number of network devices involved in service flow analysis, thereby conserving communication resources between the network devices and the network management device, as well as processing resources of each device.

[0204] The three possible scenarios described above in the embodiments of this application use a traffic analysis task indicating the analysis of a single task as an example. In practical applications, a traffic analysis task may also indicate the analysis of two or more tasks. Accordingly, the traffic analysis results may include one or more of a synchronization measurement result, a throughput performance measurement result, or a load balancing measurement result.

[0205] In addition, in the embodiment of the present application, the network management device can also count the network paths of multiple service flows of the same service. For example, the network management device can send a measurement command to all network devices in the network to instruct the network devices that receive the measurement command to report the flow identifiers of one or more flows passing through themselves. The flow identifier may include the source IP address, the destination IP address, and the QP number. The network management device can identify the same flow passing through different network devices based on the flow identifier and determine whether the flow is a service flow of the service in combination with the service-associated IP address in the measurement task information. In addition, in the scenario where the service is collective communication, it can also be combined with the synchronization measurement results of the flow to determine which flows belong to the same collective communication, thereby determining the multiple network paths of different service flows of the same service in the network.

[0206] Optionally, the network management device can perform service traffic analysis on the service under the triggering of the service platform. For example, FIG6 is a flow chart of another service traffic analysis method provided by an embodiment of the present application. As shown in FIG6, method 600 includes but is not limited to the following steps 601 to 607.

[0207] Step 601: The service platform sends a service measurement task to a network management device. The service measurement task includes measurement task information of the service. The measurement task information includes device identifiers of multiple terminal devices associated with the service.

[0208] The service includes multiple service flows, and the multiple terminal devices include sending devices and receiving devices of the multiple service flows. The explanation of the measurement task information can refer to the above step 301, and will not be repeated in this embodiment of the present application.

[0209] Optionally, taking the business as a distributed training task as an example, the measurement task information may include the task identifier, the computing resource information occupied by the task, the collective communication domain information and the traffic analysis task. Among them, the task identifier is a unique identifier used to distinguish the distributed training task, such as a universally unique identifier (UUID). The computing resource information occupied by the task includes the IP address of the computing node involved in this task, etc. The collective communication domain information includes the collective communication domain identifier of one or more collective communications involved in this task and the IP address of the computing node participating in the collective communication. The collective communication domain information may also include a collective communication mode and / or a collective communication algorithm.

[0210] Optionally, collective communication modes include, but are not limited to, AllReduce, Broadcast, AllGather, ReduceScatter, All-to-all, and Peer-to-Peer (P2P). Different collective communication modes can be used to accomplish data synchronization tasks for different purposes within distributed tasks. Collective communication algorithms include ring algorithms, halving and doubling (HD) algorithms, and tree algorithms.

[0211] Furthermore, after receiving the service measurement task sent by the service platform, the network management device may execute the following step 602 .

[0212] Step 602: The network management device sends a measurement command to multiple network devices in the network. The measurement command is used to instruct the network device that receives the measurement command to count traffic characteristics of one or more flows passing through the network device.

[0213] Optionally, the measurement command indicates the traffic characteristics to be counted. The measurement command may also include a measurement indication, which may include one or more of a source IP address set, a destination IP address set, an inbound indication, an outbound indication, or an interface indication. The measurement command may also include the effective time of the measurement task. For an explanation of the measurement command, refer to step 302 above and will not be further elaborated herein in this embodiment of the present application.

[0214] The implementation of this step 602 can refer to the implementation of the above-mentioned step 302, and the embodiment of the present application will not be repeated here.

[0215] Furthermore, after receiving the measurement command sent by the network management device, the network device in the network may execute the following steps 603 to 604 .

[0216] Step 603: The network device collects statistics on traffic characteristics of one or more flows passing through the network device according to the measurement command.

[0217] With reference to the first implementation of the first possible scenario described above, the measurement command indicates that the traffic characteristics to be counted include the start and end times of each data transmission. Accordingly, the measurement result includes a first traffic characteristic of a first flow, where the first flow is any one of one or more flows flowing through the network device, the first flow including multiple packet groups used to respectively complete multiple data transmissions, and the first traffic characteristic includes the transmission start and end times of each of the multiple packet groups.

[0218] With reference to the second implementation of the first possible scenario described above, the measurement command indicates that the traffic characteristics to be counted include the flow completion time. Accordingly, the measurement result includes the second traffic characteristic of the second flow, where the second flow is any one of the one or more flows flowing through the network device, the second flow includes multiple message groups respectively used to complete multiple data transmissions, and the second traffic characteristic includes the flow completion time of the second flow within one or more statistical time periods. The flow completion time of the second flow within each statistical time period is obtained based on the transmission duration of one or more message groups in the multiple message groups within the statistical time period at the transmission end time. The transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group.

[0219] With reference to the first implementation of the second possible scenario described above, the measurement command indicates that the traffic characteristics to be counted include the start time, end time, and flow rate of each data transmission. Accordingly, the measurement result includes third traffic characteristics of a third flow, where the third flow is any one of one or more flows flowing through the network device, the third flow includes multiple packet groups used to complete multiple data transmissions, and the third traffic characteristics include the transmission start time, transmission end time, and flow rate of each of the multiple packet groups.

[0220] With reference to the second implementation of the second possible scenario described above, the measurement command indicates that the traffic characteristics to be counted include throughput. Accordingly, the measurement result includes a fourth traffic characteristic of the fourth flow, where the fourth flow is any one of the one or more flows flowing through the network device, and the fourth flow includes multiple message groups respectively used to complete multiple data transmissions, and the fourth traffic characteristic includes the throughput of the network device for the fourth flow within one or more statistical time periods. The throughput of the fourth flow within each statistical time period is obtained based on the transmission duration and flow size of one or more message groups in the multiple message groups within the statistical time period at the end of transmission. The transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group.

[0221] With reference to the third possible scenario described above, the measurement command indicates that the traffic characteristics to be counted include traffic volume. Accordingly, the measurement result includes the traffic volume of one or more flows passing through the network device.

[0222] Step 604: The network device sends a measurement result to the network management device. The measurement result includes traffic characteristics corresponding to the flow identifiers of one or more flows passing through the network device.

[0223] The flow identifier of each flow includes the device identifier of the sending device of the flow and the device identifier of the receiving device of the flow. For the interpretation of the measurement results, reference can be made to the relevant content in the above method 300, which will not be repeated in this embodiment of the application.

[0224] Step 605: The network management device determines a traffic analysis result corresponding to the service based on the measurement results sent by the multiple network devices and the device identifications of the multiple terminal devices.

[0225] Optionally, the traffic analysis results include one or more of a synchronization measurement result, a throughput performance measurement result, or a load balancing measurement result. For interpretation of the traffic analysis results, reference may be made to the relevant content of method 300 described above. The implementation of step 605 may refer to the implementation of step 304 described above and will not be further described in detail in this embodiment of the present application.

[0226] Step 606: The network management device sends the service measurement result to the service platform, where the service measurement result includes the traffic analysis result.

[0227] Furthermore, after the service platform receives the service measurement result sent by the network management device, the following step 607 may be executed.

[0228] Step 607: The service platform outputs the traffic analysis result.

[0229] Optionally, the service platform may display the traffic analysis result, or the service platform may send the traffic analysis result to a display device for display, so that operation and maintenance personnel can view the traffic analysis result corresponding to the service to achieve quality monitoring and management of the service.

[0230] The embodiments of the present application combine the traffic characteristics of network equipment statistics with services, thereby realizing network measurement at the service granularity, which helps to monitor and analyze service quality. Taking the distributed training task or collective communication as an example, based on the characteristics of the collective communication, the traffic characteristics of each service flow in the collective communication are analyzed, the flow synchronization, throughput performance, and load balancing of the collective communication are measured, and the measurement results are reported to the service platform. This can be easily coordinated with the computing task to assist in analyzing whether the slowness or failure of distributed training is caused by the network side or the computing side.

[0231] The order of the steps in the traffic analysis method provided in the embodiments of the present application can be adjusted appropriately, and the steps can be increased or decreased accordingly. Any method that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application.

[0232] The following describes the virtual device in the embodiment of the present application by way of example.

[0233] For example, Figure 7 is a schematic diagram of the structure of a service traffic analysis device provided in an embodiment of the present application. This device can be applied to network management equipment. As shown in Figure 7, device 700 includes but is not limited to an acquisition module 701, a sending module 702, a receiving module 703, and a determination module 704. Optionally, device 700 also includes a generation module 705.

[0234] Among them, the acquisition module 701 is used to obtain the measurement task information of the service, the measurement task information includes the device identification of multiple terminal devices associated with the service, the service includes multiple service flows, and the multiple terminal devices include sending devices and receiving devices of multiple service flows. The sending module 702 is used to send measurement commands to multiple network devices in the network, and the measurement commands are used to instruct the network devices that receive the measurement commands to count the traffic characteristics of one or more flows flowing through the network devices. The receiving module 703 is used to receive measurement results sent by multiple network devices, and the measurement results sent by each network device respectively include the traffic characteristics corresponding to the flow identification of one or more flows flowing through the network device, and the flow identification of each flow includes the device identification of the sending device of the flow and the device identification of the receiving device of the flow. The determination module 704 is used to determine the traffic analysis results corresponding to the service based on the measurement results sent by multiple network devices and the device identification of multiple terminal devices.

[0235] Optionally, the service is a collective communication or a distributed training task.

[0236] Optionally, the determination module 704 is used to: for the measurement results sent by each network device, determine the business flow belonging to the business in one or more flows flowing through the network device based on the flow identifier in the measurement results and the device identifiers of multiple terminal devices; and determine the traffic analysis results based on the traffic characteristics of multiple business flows in the measurement results sent by multiple network devices.

[0237] In a first embodiment, the multiple network devices include a first network device, and the measurement results sent by the first network device include a first traffic characteristic of a first business flow, the first business flow is any business flow among the multiple business flows, and the first business flow includes multiple message groups respectively used to complete multiple data transmissions. The first traffic characteristic includes the transmission start time and transmission end time of each of the multiple message groups. The determination module 704 is used to: determine the flow completion time of the first business flow within the first statistical period based on the transmission duration of one or more message groups with transmission end times within the first statistical period among the multiple message groups, and the transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group. Among them, the traffic analysis results include the flow completion times of the multiple business flows in multiple statistical periods, and the first statistical period is any statistical period among the multiple statistical periods.

[0238] In a second embodiment, the multiple network devices include a second network device, and the measurement results sent by the second network device include a second traffic characteristic of a second business flow, the second business flow is any business flow among the multiple business flows, and the second business flow includes multiple message groups respectively used to complete multiple data transmissions. The second traffic characteristic includes the flow completion time of the second business flow within the second statistical period, and the flow completion time of the second business flow within the second statistical period is obtained based on the transmission duration of one or more message groups in the multiple message groups within the second statistical period at the end of the transmission, and the transmission duration of each message group is obtained based on the transmission end time and the transmission start time of the message group. The traffic analysis results include the flow completion time of the multiple business flows in multiple statistical periods, and the second statistical period is any statistical period among the multiple statistical periods.

[0239] Optionally, the flow completion time of a service flow in a statistical period is a statistical value of the transmission duration of all message groups in the service flow whose transmission ends within the statistical period.

[0240] Optionally, in combination with the first or second embodiment, the determination module 704 is further configured to determine, for any service flow among the multiple service flows, that the transmission quality of the service flow has deteriorated if the difference between the flow completion time of the service flow within the third statistical period and the flow completion time of the service flow within the fourth statistical period is greater than a first threshold, and the third statistical period is chronologically later than the fourth statistical period. The traffic analysis result further includes an indication of the service flow among the multiple service flows whose transmission quality has deteriorated.

[0241] In a third embodiment, the multiple network devices include a third network device, and the measurement results sent by the third network device include a third traffic characteristic of a third business flow. The third business flow is any business flow among the multiple business flows, and the third business flow includes multiple message groups respectively used to complete multiple data transmissions. The third traffic characteristic includes the transmission start time, transmission end time, and traffic size of each of the multiple message groups. The determination module 704 is used to: determine the throughput of the third business flow by the third network device within the fifth statistical period based on the transmission duration and traffic size of one or more message groups among the multiple message groups whose transmission end time is within the fifth statistical period, and the transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group. The traffic analysis results include the throughput of the business flows flowing through the multiple network devices in multiple statistical periods, and the fifth statistical period is any statistical period among the multiple statistical periods.

[0242] In a fourth embodiment, the multiple network devices include a fourth network device, and the measurement results sent by the fourth network device include the fourth traffic characteristics of the fourth business flow, the fourth business flow is any business flow among the multiple business flows, and the fourth business flow includes multiple message groups respectively used to complete multiple data transmissions. The fourth traffic characteristics include the throughput of the fourth business flow by the fourth network device within the sixth statistical period, and the throughput of the fourth business flow within the sixth statistical period is obtained based on the transmission duration and traffic size of one or more message groups in the multiple message groups whose transmission end time is within the sixth statistical period, and the transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group. Among them, the traffic analysis results include the throughput of the business flows flowing through the multiple network devices in multiple statistical periods, and the sixth statistical period is any statistical period among the multiple statistical periods.

[0243] Optionally, in combination with the third or fourth embodiment, the determination module 704 is further configured to determine, for any service flow among the multiple service flows, that the transmission quality of the service flow has deteriorated if a difference between the throughput of any network device through which the service flow passes for the service flow in a seventh statistical period and the throughput of the network device for the service flow in an eighth statistical period is greater than a second threshold, and the eighth statistical period is chronologically subsequent to the seventh statistical period. The traffic analysis result further includes an indication of the service flow among the multiple service flows whose transmission quality has deteriorated.

[0244] Optionally, in combination with any one of the first to fourth embodiments above, the sending module is used to: send measurement commands to multiple access network devices in the network respectively, and each access network device is used to connect one or more terminal devices among the multiple terminal devices to the network.

[0245] In a fifth embodiment, the measurement results sent by each network device include the flow size of one or more flows flowing through the network device. The determination module 704 is configured to: for each network device, determine the cumulative flow size of the service for the network device based on the flow sizes of all service flows in the measurement results sent by the network device; and determine the load balancing degree of the multiple network devices for the service based on the cumulative flow sizes of the multiple network devices for the service, wherein the load balancing degree is negatively correlated with the difference between the cumulative flow sizes of the multiple network devices for the service. The traffic analysis results include the load balancing degree of the multiple network devices for the service.

[0246] Optionally, in combination with the fifth embodiment, the sending module 702 is configured to send measurement commands to all network devices on the network paths of the multiple service flows in the network respectively.

[0247] Optionally, the measurement command also includes a measurement indication, which includes one or more of the following: a source IP address set, which includes the IP addresses of one or more terminal devices among a plurality of terminal devices, and the source IP address set is used to instruct the network device that receives the measurement command to count the traffic characteristics of flows passing through the network device, and whose source IP addresses belong to the source IP address set; a destination IP address set, which includes the IP addresses of one or more terminal devices among a plurality of terminal devices, and the destination IP address set is used to instruct the network device that receives the measurement command to count the traffic characteristics of flows passing through the network device, and whose destination IP addresses belong to the destination IP address set; an inbound direction indication, which is used to instruct the network device that receives the measurement command to count the traffic characteristics of one or more flows received by the network device; an outbound direction indication, which is used to instruct the network device that receives the measurement command to count the traffic characteristics of one or more flows sent by the network device; an interface indication, which is used to instruct the network device that receives the measurement command to count the traffic characteristics of flows passing through one or more specified interfaces of the network device, and the measurement result sent by the network device also includes an interface identifier, which is used to indicate the interface through which the flow passes on the network device.

[0248] Optionally, the measurement task information also includes a traffic analysis task, which indicates a traffic analysis type. A generation module 705 is configured to generate a measurement command based on the traffic analysis task, where the type of traffic features to be statistically analyzed indicated by the measurement command matches the traffic analysis type indicated by the traffic analysis task. Accordingly, a determination module is configured to determine a traffic analysis result for a service under the traffic analysis type indicated by the traffic analysis task based on the measurement results sent by the multiple network devices and the device identifiers of the multiple terminal devices.

[0249] Optionally, the acquisition module 701 is configured to receive a service measurement task sent by a service platform, where the service measurement task includes measurement task information.

[0250] Optionally, the sending module 702 is further configured to send a service measurement result to the service platform, where the service measurement result includes a traffic analysis result.

[0251] Optionally, the multiple service flows are RDMA flows.

[0252] For example, FIG8 is a schematic diagram of the structure of another service traffic analysis device provided in an embodiment of the present application. The device can be applied to a service platform. As shown in FIG8, the device 800 includes but is not limited to a sending module 801 and a receiving module 802. Optionally, the device 800 also includes an output module 803.

[0253] The sending module 801 is configured to send a service measurement task to a network management device. The service measurement task includes measurement task information for the service. The measurement task information includes device identifiers of multiple terminal devices associated with the service. The service includes multiple service flows, and the multiple terminal devices include sending devices and receiving devices for the multiple service flows. The receiving module 802 is configured to receive service measurement results sent by the network management device. The service measurement results include traffic analysis results corresponding to the service.

[0254] Optionally, the service is a collective communication or a distributed training task.

[0255] Optionally, the output module 803 is used to output traffic analysis results.

[0256] For example, FIG9 is a schematic diagram of the structure of another service flow analysis device provided in an embodiment of the present application. The device can be applied to a network device. As shown in FIG9, the device 900 includes but is not limited to a receiving module 901, a processing module 902, and a sending module 903.

[0257] Receiving module 901 is configured to receive a measurement command sent by a network management device, the measurement command instructing the network device to collect traffic characteristics of one or more flows passing through the network device. Processing module 902 is configured to collect traffic characteristics of one or more flows passing through the network device based on the measurement command. Sending module 903 is configured to send measurement results to the network management device, the measurement results including traffic characteristics corresponding to flow identifiers of one or more flows passing through the network device. Each flow identifier includes the device identifier of the sending device of the flow and the device identifier of the receiving device of the flow.

[0258] Optionally, the measurement command also includes a measurement indication, which includes one or more of the following: a source IP address set, which includes one or more IP addresses, and the source IP address set is used to instruct the network device to count the traffic characteristics of flows passing through the network device and whose source IP addresses belong to the source IP address set; a destination IP address set, which includes one or more IP addresses, and the destination IP address set is used to instruct the network device to count the traffic characteristics of flows passing through the network device and whose destination IP addresses belong to the destination IP address set; an inbound direction indication, which is used to instruct the network device to count the traffic characteristics of one or more flows received by the network device; an outbound direction indication, which is used to instruct the network device to count the traffic characteristics of one or more flows sent by the network device; an interface indication, which is used to instruct the network device to count the traffic characteristics of flows passing through one or more specified interfaces of the network device. The measurement result sent by the network device also includes an interface identifier, which is used to indicate the interface through which the flow passes on the network device.

[0259] Optionally, the measurement results include a first traffic characteristic of a first stream, where the first stream is any one or more streams, and the first stream includes multiple message groups respectively used to complete multiple data transmissions, and the first traffic characteristic includes the transmission start time and transmission end time of each of the multiple message groups.

[0260] Optionally, the measurement results include a second traffic characteristic of the second flow, where the second flow is any flow among one or more flows, and the second flow includes multiple message groups respectively used to complete multiple data transmissions. The second traffic characteristic includes the flow completion time of the second flow within one or more statistical time periods, and the flow completion time of the second flow within each statistical time period is obtained based on the transmission duration of one or more message groups in the multiple message groups within the statistical time period at the transmission end time, and the transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group.

[0261] Optionally, the measurement results include a third traffic characteristic of a third stream, where the third stream is any one or more streams, and the third stream includes multiple message groups respectively used to complete multiple data transmissions, and the third traffic characteristic includes the transmission start time, transmission end time and traffic size of each of the multiple message groups.

[0262] Optionally, the measurement results include a fourth traffic characteristic of the fourth flow, where the fourth flow is any flow among one or more flows, and the fourth flow includes multiple message groups respectively used to complete multiple data transmissions. The fourth traffic characteristic includes the throughput of the network device for the fourth flow in one or more statistical time periods, and the throughput of the fourth flow in each statistical time period is obtained based on the transmission duration and traffic size of one or more message groups in the multiple message groups at the end time of transmission in the statistical time period, and the transmission duration of each message group is obtained based on the transmission end time and transmission start time of the message group.

[0263] Optionally, the measurement result includes flow rates of one or more flows.

[0264] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0265] The following is an example of the basic hardware structure of the network device in the embodiment of the present application.

[0266] For example, Figure 10 is a schematic diagram of the hardware structure of a network device provided in an embodiment of the present application. As shown in Figure 10, network device 1000 includes a processor 1001, a forwarding chip 1002, and at least one network interface 1003. Optionally, in conjunction with Figure 2, network device 1000 in Figure 10 is any network device 203 shown in Figure 2.

[0267] Optionally, the processor 1001 includes a central processing unit (CPU) and / or a dedicated hardware chip. The CPU refers to a general-purpose CPU with high scalability and flexibility. The CPU is, for example, a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The dedicated hardware chip is a high-performance processing hardware module. The dedicated hardware chip includes at least one of an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a network processor (NP).

[0268] The forwarding chip 1002 is used to forward requests and data. For example, the forwarding chip 1002 is used by the network device 1000 to forward access request messages to the protected device and to forward access response messages to the host.

[0269] At least one network interface 1003 includes, for example, network interface 1, network interface 2, network interface 3, ..., network interface n in FIG10 . The network interface 1003 uses any transceiver-like device for communicating with other devices or communication networks. For example, the network interface 1 in FIG10 communicates with the terminal device, and the network interface 2 in FIG10 communicates with the network management device. Optionally, the network interface 1003 includes at least one of a wired network interface or a wireless network interface. The wired network interface is, for example, an Ethernet interface. The Ethernet interface is, for example, an optical interface, an electrical interface, or a combination thereof. The wireless network interface is, for example, a wireless local area network (WLAN) interface, a cellular network interface, or a combination thereof.

[0270] At least one network interface 1003 is connected to the forwarding chip 1002, and the forwarding chip 1002 is connected to the processor 1001 via an internal connection 1004. The internal connection 1004 includes a path for transmitting data between the network interface 1003, the forwarding chip 1002, and the processor 1001. Optionally, the internal connection 1004 is a single board or a bus. For example, the internal connection 1004 is Ethernet, fiber channel, PCI-E (peripheral component interconnect express, PCI Express, a high-speed serial computer bus), RapidIO (a high-performance, low-pin-count, packet-switching-based interconnect architecture), InfiniBand, or a XAUI bus (an interface extender characterized by connecting the Ethernet media access control (MAC) layer to the physical layer).

[0271] Optionally, network device 1000 further includes a content addressable memory (CAM) 1005. CAM 1005 is, for example, a ternary content addressable memory (TCAM). Optionally, CAM 1005 exists independently and is connected to forwarding chip 1002 via the aforementioned internal connection 1004. Alternatively, CAM 1005 and forwarding chip 1002 are integrated, i.e., CAM 1005 functions as memory within forwarding chip 1002.

[0272] Optionally, the network device 1000 further includes a memory 1006. The memory 1006 is, for example, a read-only memory (ROM) or other type of static storage device that can store static information and instructions, or a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code 10010 in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 1006 is, for example, independent and connected to the processor 1001 via the internal connection 1004. Alternatively, the memory 1006 and the processor 1001 are integrated together.

[0273] The memory 1006 stores an operating system 1007 and program code 10010. Optionally, the processor 1001 reads the operating system 1007 from the memory 1006 and runs the operating system 1007. The processor 1001 also reads the program code 10010 from the memory 1006, and implements the actions performed by the network device in the above method provided in the embodiment of the present application by running the program code 10010 on the operating system 1007.

[0274] Optionally, the above-mentioned devices are respectively provided on independent chips, or at least partially or entirely provided on the same chip. Whether each device is provided independently on different chips or integrated on one or more chips often depends on the needs of product design. The embodiments of the present application do not limit the specific implementation of the above-mentioned devices.

[0275] The following is an example of the basic hardware structure of the network management device / service platform in the embodiment of the present application.

[0276] For example, Figure 11 is a schematic diagram of the hardware structure of a network management device / service platform provided in an embodiment of the present application. As shown in Figure 11, the network management device / service platform 1100 includes a processor 1101 and a memory 1102, and the processor 1101 and the memory 1102 are connected via a bus 1103. Figure 11 illustrates the processor 1101 and the memory 1102 as independent of each other. Optionally, the processor 1101 and the memory 1102 are integrated together. Optionally, in conjunction with Figure 2, the network management device / service platform 1100 in Figure 11 can be the network management device 201 or the service platform 204 shown in Figure 2.

[0277] Memory 1102 is used to store computer programs, including operating systems and program code. Memory 1102 can be any type of storage medium, such as ROM, RAM, EEPROM, CD-ROM, flash memory, optical storage, registers, optical disk storage, optical disc storage, magnetic disk, or other magnetic storage device.

[0278] Processor 1101 is a general-purpose processor or a dedicated processor. Processor 1101 may be a single-core processor or a multi-core processor. Processor 1101 includes at least one circuit to execute the actions performed by the network management device / service platform in the above method provided in the embodiment of the present application.

[0279] Optionally, the network management device / service platform 1100 further includes a network interface 1104, which is connected to the processor 1101 and the memory 1102 via the bus 1103. The network interface 1104 enables the network management device / service platform 1100 to communicate with other devices.

[0280] Optionally, the network management device / service platform 1100 further includes an input / output (I / O) interface 1105, which is connected to the processor 1101 and the memory 1102 via the bus 1103. The processor 1101 can receive input commands or data through the I / O interface 1105. The I / O interface 1105 is used to connect the network management device / service platform 1100 to input devices such as a keyboard and a mouse. Optionally, in some possible scenarios, the network interface 1104 and the I / O interface 1105 are collectively referred to as a communication interface.

[0281] Optionally, network management device / service platform 1100 further includes a display 1106, which is connected to processor 1101 and memory 1102 via bus 1103. Display 1106 can be used to display intermediate and / or final results generated by processor 1101 executing the above-described method, such as traffic analysis results corresponding to the service. In one possible implementation, display 1106 is a touchscreen display to provide a human-computer interaction interface.

[0282] The bus 1103 is any type of communication bus used to interconnect the internal components of the network management device / service platform 1100, such as a system bus. The embodiments of the present application illustrate the example of the aforementioned components within the network management device / service platform 1100 being interconnected via the bus 1103. Alternatively, the aforementioned components within the network management device / service platform 1100 may be communicatively connected to each other using other connection methods besides the bus 1103, such as interconnecting the aforementioned components within the network management device / service platform 1100 via a logical interface within the network management device / service platform 1100.

[0283] The above-mentioned devices can be provided on separate chips, or at least partially or entirely on the same chip. Whether to provide each device independently on different chips or to integrate them on one or more chips often depends on the product design requirements. The embodiments of this application do not limit the specific implementation of the above-mentioned devices.

[0284] The network management device / service platform 1100 shown in FIG11 is merely exemplary. During implementation, the network management device / service platform 1100 includes other components, which are not listed here. The network management device / service platform 1100 shown in FIG11 can analyze service traffic by executing all or part of the steps of the method provided in the above embodiment.

[0285] The present application also provides a service traffic analysis system, including a network management device and multiple network devices. The network management device is configured to execute the actions executed by the network management device in the above method embodiment. The network devices are configured to execute the actions executed by the network devices in the above method embodiment.

[0286] Optionally, the system further includes a service platform, and the service platform is used to execute the actions executed by the service platform in the above method embodiment.

[0287] An embodiment of the present application further provides a computer-readable storage medium having instructions stored thereon. When the instructions are executed by a processor, the actions performed by the network management device, network device or service platform in the above method embodiment are implemented.

[0288] An embodiment of the present application further provides a computer program product, including a computer program. When the computer program is executed by a processor, the computer program implements the actions performed by the network management device, network device or service platform in the above method embodiment.

[0289] An embodiment of the present application further provides a chip, which includes a programmable logic circuit and / or program instructions. When the chip is running, it implements the actions performed by the network management device, network device or service platform in the above method embodiment.

[0290] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0291] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.

[0292] In the embodiments of the present application, the terms “first”, “second” and “third” are used for descriptive purposes only and should not be understood as indicating or implying relative importance.

[0293] In this application, the term "and / or" simply describes an association between related objects, indicating that three possible relationships exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the related objects are in an "or" relationship.

[0294] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0295] The above description is merely an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the concepts and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A service traffic analysis method, characterized in that, The method includes: Obtaining measurement task information of a service, where the measurement task information includes device identifiers of multiple terminal devices associated with the service, the service includes multiple service flows, and the multiple terminal devices include sending devices and receiving devices of the multiple service flows; Sending measurement commands to multiple network devices in the network, where the measurement commands are used to instruct the network devices that receive the measurement commands to count traffic characteristics of one or more flows passing through the network devices; Receiving measurement results sent by the multiple network devices, where the measurement results sent by each network device respectively include traffic characteristics corresponding to flow identifiers of one or more flows passing through the network device, and the flow identifier of each flow includes the device identifier of the sending device of the flow and the device identifier of the receiving device of the flow; Determining a traffic analysis result corresponding to the service according to the measurement results sent by the multiple network devices and the device identifiers of the multiple terminal devices.

2. The method according to claim 1, characterized in that, The service is a group communication or a distributed training task.

3. The method according to claim 1 or 2, characterized in that, The determining the traffic analysis result corresponding to the service according to the measurement results sent by the multiple network devices and the device identifiers of the multiple terminal devices includes: For the measurement result sent by each network device, determining, according to the flow identifier in the measurement result and the device identifiers of the multiple terminal devices, service flows belonging to the service among the one or more flows passing through the network device; Determining the traffic analysis result according to the traffic characteristics of the multiple service flows in the measurement results sent by the multiple network devices.

4. The method according to claim 3, characterized in that, The multiple network devices include a first network device, and the measurement result sent by the first network device includes a first traffic characteristic of a first service flow, the first service flow being any one of the multiple service flows, and the first service flow includes multiple packet groups respectively used to complete multiple data transmissions; The first traffic characteristic includes the transmission start time and transmission end time of each of the multiple packet groups, and the determining the traffic analysis result according to the traffic characteristics of the multiple service flows in the measurement results sent by the multiple network devices includes: Determining the flow completion time of the first service flow in the first statistical period according to the transmission durations of one or more packet groups whose transmission end times are within the first statistical period among the multiple packet groups, where the transmission duration of each packet group is obtained based on the transmission end time and transmission start time of the packet group; Wherein, the traffic analysis result includes the flow completion times of the multiple service flows in multiple statistical periods respectively, and the first statistical period is any one of the multiple statistical periods.

5. The method according to claim 3, wherein The multiple network devices include a second network device, and the measurement result sent by the second network device includes a second traffic characteristic of a second service flow, the second service flow being any one of the multiple service flows, and the second service flow includes multiple packet groups respectively used to complete multiple data transmissions; The second traffic characteristic includes the flow completion time of the second traffic flow within the second statistical period. The flow completion time of the second traffic flow within the second statistical period is obtained based on the transmission durations of one or more packet groups among the multiple packet groups whose transmission end times are within the second statistical period. The transmission duration of each packet group is obtained based on the transmission end time and the transmission start time of the packet group; Wherein, the traffic analysis result includes the flow completion times of the multiple traffic flows within multiple statistical periods respectively, and the second statistical period is any one of the multiple statistical periods.

6. The method according to claim 4 or 5, characterized in that, The flow completion time of a traffic flow within a statistical period is the statistical value of the transmission durations of all the packet groups in the traffic flow whose transmission end times are within the statistical period.

7. The method according to any one of claims 4 to 6, characterized in that, The method further includes: For any one of the multiple traffic flows, if the difference between the flow completion time of the traffic flow within the third statistical period and the flow completion time of the traffic flow within the fourth statistical period is greater than a first threshold, it is determined that the transmission quality of the traffic flow has deteriorated, and the third statistical period is temporally after the fourth statistical period; Wherein, the traffic analysis result further includes an indication of the traffic flows among the multiple traffic flows whose transmission qualities have deteriorated.

8. The method according to claim 3, wherein The multiple network devices include a third network device. The measurement result sent by the third network device includes the third traffic characteristic of a third traffic flow, where the third traffic flow is any one of the multiple traffic flows, and the third traffic flow includes multiple packet groups respectively used to complete multiple data transmissions; The third traffic characteristic includes the transmission start time, transmission end time, and traffic volume of each of the multiple packet groups. Determining the traffic analysis result according to the traffic characteristics of the multiple traffic flows in the measurement results sent by the multiple network devices includes: Determining the throughput of the third network device for the third traffic flow within the fifth statistical period according to the transmission durations and traffic volumes of one or more packet groups among the multiple packet groups whose transmission end times are within the fifth statistical period. The transmission duration of each packet group is obtained based on the transmission end time and the transmission start time of the packet group; Wherein, the traffic analysis result includes the throughputs of the multiple network devices for the traffic flows passing through themselves within multiple statistical periods respectively, and the fifth statistical period is any one of the multiple statistical periods.

9. The method according to claim 3, wherein The multiple network devices include a fourth network device. The measurement result sent by the fourth network device includes the fourth traffic characteristic of a fourth traffic flow, where the fourth traffic flow is any one of the multiple traffic flows, and the fourth traffic flow includes multiple packet groups respectively used to complete multiple data transmissions; The fourth traffic characteristic includes the throughput of the fourth service flow by the fourth network device within the sixth statistical period. The throughput of the fourth service flow within the sixth statistical period is obtained based on the transmission durations and traffic volumes of one or more packet groups among the multiple packet groups whose transmission end times are within the sixth statistical period. The transmission duration of each packet group is obtained based on the transmission end time and the transmission start time of the packet group; Among them, the traffic analysis result includes the throughputs of the service flows passing through each of the multiple network devices within multiple statistical periods, and the sixth statistical period is any one of the multiple statistical periods.

10. The method according to claim 8 or 9, characterized in that, The method further includes: For any one of the multiple service flows, if the difference between the throughput of the service flow by any network device through which the service flow passes within the seventh statistical period and the throughput of the service flow by the network device within the eighth statistical period is greater than a second threshold, it is determined that the transmission quality of the service flow has deteriorated, and the eighth statistical period is chronologically after the seventh statistical period; Among them, the traffic analysis result further includes an indication of the service flows among the multiple service flows whose transmission quality has deteriorated.

11. According to the method described in any one of claims 4 to 10, characterized in that, The sending of measurement commands to multiple network devices in the network includes: Sending the measurement commands to multiple access network devices in the network respectively, and each access network device is respectively used to connect one or more of the multiple terminal devices to the network.

12. The method according to claim 3, wherein The measurement results sent by each network device respectively include the traffic volumes of one or more flows passing through the network device. The determining of the traffic analysis result according to the traffic characteristics of the multiple service flows in the measurement results sent by the multiple network devices includes: For each network device, determining the cumulative traffic volume of the network device for the service according to the traffic volumes of all service flows in the measurement results sent by the network device; Determining the load balancing degree of the multiple network devices for the service according to the cumulative traffic volumes of the multiple network devices for the service respectively, and the load balancing degree is negatively correlated with the difference between the cumulative traffic volumes of the multiple network devices for the service; Among them, the traffic analysis result includes the load balancing degree of the multiple network devices for the service.

13. The method according to claim 12, wherein The sending of measurement commands to multiple network devices in the network includes: Sending the measurement commands to all network devices on the network path of the multiple service flows in the network respectively.

14. The method according to any one of claims 1 to 13, characterized in that, The measurement command further includes a measurement indication, and the measurement indication includes one or more of the following: A set of source Internet Protocol (IP) addresses, where the set of source IP addresses includes the IP addresses of one or more of the multiple terminal devices, and the set of source IP addresses is used to instruct the network device that receives the measurement command to count the traffic characteristics of the flows passing through the network device and whose source IP addresses belong to the set of source IP addresses; A set of destination IP addresses, where the set of destination IP addresses includes the IP addresses of one or more of the multiple terminal devices, and the set of destination IP addresses is used to instruct a network device that receives the measurement command to count the traffic characteristics of the flows that pass through the network device and whose destination IP addresses belong to the set of destination IP addresses; An incoming direction indication, where the incoming direction indication is used to instruct a network device that receives the measurement command to count the traffic characteristics of one or more flows received by the network device; An outgoing direction indication, where the outgoing direction indication is used to instruct a network device that receives the measurement command to count the traffic characteristics of one or more flows sent by the network device; An interface indication, where the interface indication is used to instruct a network device that receives the measurement command to count the traffic characteristics of the flows passing through one or more specified interfaces of the network device, and the measurement result sent by the network device also includes an interface identifier, and the interface identifier is used to indicate the interface through which the flow passes on the network device.

15. The method according to any one of claims 1 to 14, characterized in that, The measurement task information further includes a traffic analysis task, and the traffic analysis task is used to indicate the traffic analysis type. The method further includes: Generating the measurement command according to the traffic analysis task, and the type of the traffic characteristics indicated to be counted by the measurement command matches the traffic analysis type indicated by the traffic analysis task; The determining the traffic analysis result corresponding to the service according to the measurement results sent by the multiple network devices and the device identifiers of the multiple terminal devices includes: Determining the traffic analysis result of the service under the traffic analysis type indicated by the traffic analysis task according to the measurement results sent by the multiple network devices and the device identifiers of the multiple terminal devices.

16. The method according to any one of claims 1 to 15, characterized in that, The obtaining the measurement task information of the service includes: Receiving a service measurement task sent by a service platform, and the service measurement task includes the measurement task information.

17. The method according to claim 16, wherein The method further includes: Sending a service measurement result to the service platform, and the service measurement result includes the traffic analysis result.

18. A method for analyzing service traffic, characterized in that The method includes: Sending a service measurement task to a network management device, where the service measurement task includes the measurement task information of the service, the measurement task information includes the device identifiers of multiple terminal devices associated with the service, the service includes multiple service flows, and the multiple terminal devices include the sending devices and receiving devices of the multiple service flows; Receiving the service measurement result sent by the network management device, and the service measurement result includes the traffic analysis result corresponding to the service.

19. The method according to claim 18, characterized in that, The service is a group communication or a distributed training task.

20. The method according to claim 18 or 19, characterized in that, The method further includes: Outputting the traffic analysis result.

21. A business traffic analysis method, characterized in that, Applied to a network device, the method includes: Receiving a measurement command sent by a network management device, and the measurement command is used to instruct the network device to count the traffic characteristics of one or more flows passing through the network device; Counting the traffic characteristics of one or more flows passing through the network device according to the measurement command; Send the measurement result to the network management device, where the measurement result includes the traffic characteristics corresponding to the flow identifiers of one or more flows passing through the network device, and the flow identifier of each flow includes the device identifier of the sending device of the flow and the device identifier of the receiving device of the flow.

22. The method according to claim 21, wherein The measurement command further includes a measurement indication, and the measurement indication includes one or more of the following: A set of source Internet Protocol (IP) addresses, where the set of source IP addresses includes one or more IP addresses, and the set of source IP addresses is used to instruct the network device to count the traffic characteristics of the flows passing through the network device and having source IP addresses belonging to the set of source IP addresses; A set of destination IP addresses, where the set of destination IP addresses includes one or more IP addresses, and the set of destination IP addresses is used to instruct the network device to count the traffic characteristics of the flows passing through the network device and having destination IP addresses belonging to the set of destination IP addresses; An inbound direction indication, which is used to instruct the network device to count the traffic characteristics of one or more flows received by the network device; An outbound direction indication, which is used to instruct the network device to count the traffic characteristics of one or more flows sent by the network device; An interface indication, which is used to instruct the network device to count the traffic characteristics of the flows passing through one or more specified interfaces of the network device, and the measurement result sent by the network device further includes an interface identifier, where the interface identifier is used to indicate the interface through which the flow passes on the network device.

23. A service traffic analysis device, characterized in that The apparatus includes a plurality of functional modules, and the plurality of functional modules interact with each other to implement the method according to any one of claims 1 to 22.

24. A service traffic analysis device, characterized in that, including: a processor and a memory; The memory is used to store a computer program, and the computer program includes program instructions; The processor is used to call the computer program to implement the method according to any one of claims 1 to 22.

25. A service traffic analysis system, characterized in that, including: a network management device and a plurality of network devices, where the network management device is used to execute the method according to any one of claims 1 to 17, and the network device is used to execute the method according to claim 21 or 22.

26. The system according to claim 25, wherein The system further includes a service platform, and the service platform is used to execute the method according to any one of claims 18 to 20.

27. A computer-readable storage medium, characterized in that, Instructions are stored on the computer-readable storage medium, and when the instructions are executed by the processor, the method according to any one of claims 1 to 22 is implemented.

28. A computer program product, characterized in that, including a computer program, and when the computer program is executed by the processor, the method according to any one of claims 1 to 22 is implemented.

Images