Congestion processing method and device and related equipment
By monitoring the port queue buffer utilization through the switching chip and sending it to the CPU for analysis, combined with historical traffic information, the problems of high detection latency and inaccurate positioning in traditional methods are solved, achieving fast and accurate congestion detection and diagnosis, and improving the efficiency and stability of large-scale distributed training.
Patent Information
- Application Number
- CN202511028788.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-24
- Publication Date
- 2025-10-03
AI Technical Summary
Traditional network congestion detection methods have high detection latency, inaccurate positioning, and lack of in-depth analysis. They make it difficult to quickly discover congested ports and identify traffic components, affecting the efficiency and stability of large-scale distributed training.
The port queue buffer utilization is monitored by the switching chip. When the preset high threshold is reached, the target traffic message is sent to the CPU for analysis. Combined with the historical traffic analysis results, the cause of congestion is determined, including traffic component information such as source IP, destination IP, protocol type and packet size.
Achieve second-level congestion detection, accurately locate the source of congestion, reduce troubleshooting time and cost, improve network resource utilization, and meet the real-time requirements of large model training.
Smart Images
Figure CN120750855A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data processing technology, and in particular to a congestion processing method, apparatus, and related equipment. Background Art
[0002] With the rapid development of deep learning technology, large-scale model training has placed higher demands on network performance. In large-scale distributed training, network congestion is one of the key factors affecting training efficiency. Traditional network congestion detection methods are usually based on indicators such as end-to-end latency and packet loss rate, and have the following problems:
[0003] High detection latency: Traditional methods rely on end-to-end performance indicators and cannot capture congestion within the network in real time, resulting in high detection latency.
[0004] Inaccurate positioning: Traditional methods make it difficult to accurately locate the specific location of congestion (such as switch ports, queues, etc.), increasing the difficulty of troubleshooting.
[0005] Lack of in-depth analysis: Traditional methods cannot conduct in-depth analysis of traffic components during congestion, making it difficult to identify the root cause of congestion.
[0006] Therefore, there is an urgent need for a method that can quickly discover congested ports, queue information, and analyze congested traffic components to improve the efficiency and stability of large model training. Summary of the Invention
[0007] The present application provides a congestion handling method, apparatus, and related equipment.
[0008] In a first aspect, the present application provides a congestion handling method, applied to a network device; the method comprises:
[0009] The switching chip of the network device monitors the utilization rate of each port queue buffer;
[0010] If the switching chip determines that the target port queue buffer utilization reaches a preset high threshold, the target traffic message currently carried by the target port queue is sent to the CPU of the network device;
[0011] The CPU analyzes the target traffic message to obtain the target traffic analysis result;
[0012] The CPU sends the target traffic analysis result to the analyzer, so that the analyzer determines a cause of congestion of the target port queue based on the target traffic analysis result and a historical traffic analysis result of the target port queue.
[0013] Optionally, if the switching chip determines that the utilization rate of the target port queue buffer reaches a preset high threshold, the step of sending the target traffic message currently carried by the target port queue to the CPU includes:
[0014] If the switching chip determines that the utilization rate of the target port queue buffer reaches a preset high threshold, the port state of the target port is marked as a burst state;
[0015] The switching chip triggers a congestion event report and starts a message copy function to copy the target traffic message carried by the target port queue and send it to the CPU.
[0016] Optionally, the step of the CPU analyzing the target traffic message to obtain the target traffic analysis result includes:
[0017] The CPU analyzes the traffic component information of the target traffic based on the target traffic message, wherein the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow.
[0018] In a second aspect, the present application provides a congestion handling method, applied to an analyzer, the method comprising:
[0019] receiving a target traffic analysis result sent by a network device, wherein, when determining that a utilization rate of a target port queue buffer reaches a preset high threshold, the switching chip of the network device sends the target traffic message currently carried by the target port queue to a CPU of the network device; the CPU analyzes the target traffic message to obtain the target traffic analysis result;
[0020] Based on the historical traffic analysis result of the target port queue and the target traffic analysis result, a cause of congestion of the target port queue is determined.
[0021] Optionally, the target traffic analysis result includes traffic component information of the target traffic, and the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow;
[0022] The step of determining a cause of congestion of the target port queue based on the target traffic analysis result and a historical traffic analysis result of the target port queue includes:
[0023] Based on the source IP, destination IP, protocol type, and packet size of each target flow, as well as the source IP, destination IP, protocol type, and packet size of each flow included in the historical traffic analysis results, determine whether there is a newly added first flow in the traffic carried by the target port queue, and determine whether there is a second flow with an increase in traffic bandwidth greater than a set threshold;
[0024] If it is determined that the first flow and / or the second flow exist, the first flow and / or the second flow are determined as the cause of congestion of the target port queue.
[0025] In a third aspect, the present application provides a network device, comprising a switching chip and a processor:
[0026] The switch chip is used to monitor the utilization of each port queue buffer;
[0027] If the switching chip determines that the target port queue buffer utilization reaches a preset high threshold, the target traffic message currently carried by the target port queue is sent to the processor of the network device;
[0028] The processor is configured to analyze the target traffic message to obtain the target traffic analysis result;
[0029] The processor is further configured to send the target traffic analysis result to an analyzer, so that the analyzer determines a cause of congestion of the target port queue based on the target traffic analysis result and a historical traffic analysis result of the target port queue.
[0030] Optionally, if the switching chip determines that the utilization rate of the target port queue buffer reaches a preset high threshold, when sending the target traffic message currently carried by the target port queue to the processor, the switching chip is specifically configured to:
[0031] If it is determined that the target port queue buffer utilization reaches a preset high threshold, the port state of the target port is marked as a burst state;
[0032] The congestion event report is triggered, and the message copy function is started to copy the target traffic message carried by the target port queue and send it to the processor.
[0033] Optionally, when analyzing the target traffic message to obtain the target traffic analysis result, the processor is specifically configured to:
[0034] Based on the target traffic message, the traffic component information of the target traffic is analyzed, wherein the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow.
[0035] In a fourth aspect, the present application provides a congestion processing device, applied to an analyzer, the device comprising:
[0036] a receiving unit, configured to receive a target traffic analysis result sent by a network device, wherein the switching chip of the network device sends the target traffic message currently carried by the target port queue to a CPU of the network device when determining that the utilization rate of the target port queue buffer reaches a preset high threshold; the CPU analyzes the target traffic message to obtain the target traffic analysis result;
[0037] The determining unit is configured to determine a cause of congestion of the target port queue based on a historical traffic analysis result of the target port queue and the target traffic analysis result.
[0038] Optionally, the target traffic analysis result includes traffic component information of the target traffic, and the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow;
[0039] When determining a cause of congestion of the target port queue based on the target traffic analysis result and a historical traffic analysis result of the target port queue, the determining unit is specifically configured to:
[0040] Based on the source IP, destination IP, protocol type, and packet size of each target flow, as well as the source IP, destination IP, protocol type, and packet size of each flow included in the historical traffic analysis results, determine whether there is a newly added first flow in the traffic carried by the target port queue, and determine whether there is a second flow with an increase in traffic bandwidth greater than a set threshold;
[0041] If it is determined that the first flow and / or the second flow exist, the first flow and / or the second flow are determined as the cause of congestion of the target port queue.
[0042] In a fifth aspect, an embodiment of the present application provides a congestion processing device, the congestion processing device comprising:
[0043] a memory for storing program instructions;
[0044] The processor is configured to call the program instructions stored in the memory and execute the steps of the method as described in any one of the first aspects above according to the obtained program instructions.
[0045] In a sixth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable the computer to execute the steps of the method described in any one of the above-mentioned first aspects.
[0046] In a seventh aspect, an embodiment of the present application provides a congestion processing device, the congestion processing device comprising:
[0047] a memory for storing program instructions;
[0048] The processor is used to call the program instructions stored in the memory and execute the steps of the method as described in any one of the second aspects according to the obtained program instructions.
[0049] In an eighth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable the computer to execute the steps of the method described in any one of the above-mentioned second aspects.
[0050] In summary, the congestion handling method provided in the embodiment of the present application is applied to a network device; the method includes: the switching chip of the network device monitors the utilization rate of each port queue buffer; if the switching chip determines that the utilization rate of the target port queue buffer reaches a preset high threshold, the target traffic message currently carried by the target port queue is sent to the CPU of the network device; the CPU analyzes the target traffic message to obtain the target traffic analysis result; the CPU sends the target traffic analysis result to the analyzer, so that the analyzer determines the cause of the congestion of the target port queue based on the target traffic analysis result and the historical traffic analysis result of the target port queue.
[0051] The congestion handling method provided in the embodiment of the present application is adopted to achieve second-level congestion detection through the switch port queue monitoring function, significantly reducing detection delays and meeting the real-time requirements of large-scale model training for network performance. When a congestion event occurs, the traffic component information of the congested port is collected and reported, and accurately located in combination with the network topology information, reducing the time and cost of troubleshooting. A historical traffic profile of the port is generated and compared with the traffic component at the time of congestion for analysis, intelligently diagnosing the root cause of the congestion and providing targeted solutions. By accurately locating the source of congestion and intelligently diagnosing the cause of congestion, the time and resource consumption of network troubleshooting are reduced, operation and maintenance costs are reduced, and the utilization rate of network resources is improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments of the present application or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings of the embodiments of the present application.
[0053] Figure 1 A detailed flowchart of a congestion handling method provided in an embodiment of the present application;
[0054] Figure 2 A detailed flowchart of another congestion handling method provided in an embodiment of the present application;
[0055] Figure 3 A schematic diagram of the structure of a network device provided in an embodiment of the present application;
[0056] Figure 4 A schematic diagram of the structure of a congestion handling device provided in an embodiment of the application;
[0057] Figure 5 A schematic diagram of the hardware architecture of a congestion handling device provided in an embodiment of the present application;
[0058] Figure 6 A schematic diagram of the hardware architecture of another congestion handling device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0059] The terms used in the embodiments of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms "a," "the," and "the" used in this application and claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to any or all possible combinations of one or more associated listed items.
[0060] It should be understood that although the terms first, second, third, etc. may be used to describe various information in the embodiments of the present application, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" used may also be interpreted as "at the time of" or "when" or "in response to determining".
[0061] For example, see Figure 1 FIG. 1 is a detailed flow chart of a congestion handling method provided in an embodiment of the present application. The method is applied to a network device and includes the following steps:
[0062] Step 100: The switch chip of the network device monitors the utilization rate of each port queue buffer.
[0063] In this embodiment, a queue monitoring function is pre-configured on the network device port. The network device's switch chip monitors the queue buffer utilization of each port in real time. Queue buffer utilization thresholds (preset upper and lower thresholds) are pre-set. When the preset upper threshold is exceeded, a congestion event reporting analyzer is triggered, and traffic packets currently carried by the port queue are copied and reported to the network device's CPU for analysis.
[0064] Step 110: If the switching chip determines that the target port queue buffer utilization reaches a preset high threshold, the target traffic message currently carried by the target port queue is sent to the CPU of the network device.
[0065] In an embodiment of the present application, if the switching chip determines that the utilization rate of the target port queue buffer reaches a preset high threshold, then when sending the target traffic message currently carried by the target port queue to the CPU, a preferred implementation method is:
[0066] If the switching chip determines that the utilization rate of the target port queue buffer reaches a preset high threshold, the port state of the target port is marked as a burst state; the switching chip triggers a congestion event report and starts a message copy function to copy the target traffic message carried by the target port queue to the CPU.
[0067] Step 120: The CPU analyzes the target traffic message to obtain the target traffic analysis result.
[0068] In the embodiment of the present application, when the CPU analyzes the target traffic message and obtains the target traffic analysis result, a preferred implementation method is:
[0069] The CPU analyzes the traffic component information of the target traffic based on the target traffic message, wherein the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow.
[0070] In actual applications, the switching chip monitors the queue buffer utilization of the port in real time and compares it with a preset high threshold and a preset low threshold. When the queue buffer utilization reaches the preset high threshold for the first time, the switching chip determines that a port state switch has occurred and changes the port state from a normal state to a burst state. At the same time, it triggers a congestion event report and starts a message copy function, copying the message output by the port queue to the CPU of the network device for analysis. Specifically, during the burst state, if the queue buffer utilization reaches the preset high threshold again, the current state is maintained unchanged, the switching chip continues to perform the message copy operation and reports the analyzed traffic component information to the analyzer, the traffic component information includes characteristics such as source IP, destination IP, protocol type and data packet size. When the port queue buffer utilization drops to the preset low threshold, the switching chip determines a state switch and restores the port state from a burst state to a normal state, triggers an event report and stops the message copy function, and ends the current abnormality analysis cycle.
[0071] Step 130: The CPU sends the target traffic analysis result to the analyzer, so that the analyzer determines the cause of congestion of the target port queue based on the target traffic analysis result and the historical traffic analysis result of the target port queue.
[0072] For example, see Figure 2 FIG. 1 is a detailed flow chart of a congestion handling method provided in an embodiment of the present application. The method is applied to an analyzer and includes the following steps:
[0073] Step 200: Receive target traffic analysis results sent by a network device.
[0074] In an embodiment of the present application, when the switching chip of the network device determines that the utilization rate of the target port queue buffer reaches a preset high threshold, the target traffic message currently carried by the target port queue is sent to the CPU of the network device; the CPU analyzes the target traffic message to obtain the target traffic analysis result.
[0075] Step 210: Based on the historical traffic analysis result of the target port queue and the target traffic analysis result, determine the cause of congestion of the target port queue.
[0076] In an embodiment of the present application, the target traffic analysis result includes traffic component information of the target traffic, and the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow.
[0077] In other words, the analyzer also maintains the traffic analysis results of the target port queue reported in the past. Therefore, the analyzer can compare the target traffic analysis results reported for the target port queue with the traffic profiling results (e.g., the traffic analysis results of the target port queue reported last time) to determine the cause of congestion in the target port queue.
[0078] Specifically, when determining the cause of congestion of the target port queue based on the target traffic analysis result and the historical traffic analysis result of the target port queue, a preferred implementation method is:
[0079] Based on the source IP, destination IP, protocol type and packet size of each target flow, as well as the source IP, destination IP, protocol type and packet size of each flow included in the historical traffic analysis results, determine whether there is a newly added first traffic in the traffic carried by the target port queue, and determine whether there is a second traffic with an increase in traffic bandwidth greater than a set threshold; if it is determined that the first traffic and / or the second traffic exist, determine the first traffic and / or the second traffic as the cause of congestion of the target port queue.
[0080] In actual applications, the analyzer will collect historical traffic data of the switch port based on congestion events reported by network devices and traffic analysis results, and generate a traffic profile for each port, including traffic composition information, traffic path, traffic bandwidth, traffic size and other indicators at historical moments.
[0081] After receiving a congestion event, the analyzer extracts the congestion location information (network device + port + queue) and the traffic composition information at the time of congestion. This is then compared with the port's historical traffic profile to identify changes in traffic composition (e.g., sudden traffic increases, protocol type changes, etc.). Combining network topology, traffic path information, and other information, the analyzer identifies the cause of the congestion (e.g., network link failure, optical module failure, uneven traffic hashing, etc.).
[0082] For example, the congestion event report currently received carries the traffic composition information at the time of congestion as shown in Table 1:
[0083]
[0084] Table 1
[0085] At the same time, the traffic profiling function is used to query the information before the port congestion, as shown in Table 2:
[0086]
[0087] Table 2
[0088] After comparative analysis, we pinpointed the port congestion to the presence of an additional RoCE flow from 1.1.31.1 (5345) to 1.1.23.1 (4791). We then reviewed the original information for this flow in the AI traffic profile and discovered that it was originally sent from FourHundredGige1 / 0 / 64 on Leaf1 to Spine1. However, due to an optical module failure on FourHundredGige1 / 0 / 64, the link went down, causing traffic on this port to be switched to FourHundredGige1 / 0 / 62, resulting in congestion.
[0089] For example, see Figure 3 FIG. 1 is a schematic diagram of a network device according to an embodiment of the present invention, wherein the network device includes:
[0090] The switching chip 30 is used to monitor the utilization of each port queue buffer;
[0091] If the switching chip 30 determines that the target port queue buffer utilization reaches a preset high threshold, the target traffic message currently carried by the target port queue is sent to the processor 31 of the network device;
[0092] The processor 31 is configured to analyze the target traffic message to obtain the target traffic analysis result;
[0093] The processor 31 is further configured to send the target traffic analysis result to the analyzer, so that the analyzer determines a cause of congestion of the target port queue based on the target traffic analysis result and a historical traffic analysis result of the target port queue.
[0094] Optionally, if the switching chip 30 determines that the utilization rate of the target port queue buffer reaches a preset high threshold, when sending the target traffic message currently carried by the target port queue to the processor, the switching chip 30 is specifically configured to:
[0095] If it is determined that the target port queue buffer utilization reaches a preset high threshold, the port state of the target port is marked as a burst state;
[0096] The congestion event report is triggered, and the message copy function is started to copy the target traffic message carried by the target port queue and send it to the processor 31.
[0097] Optionally, when analyzing the target traffic message and obtaining the target traffic analysis result, the processor 31 is specifically configured to:
[0098] Based on the target traffic message, the traffic component information of the target traffic is analyzed, wherein the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow.
[0099] For example, see Figure 4 FIG. 1 is a schematic diagram of the structure of a congestion processing device provided in an embodiment of the present application. The device is applied to an analyzer and includes:
[0100] The receiving unit 40 is configured to receive a target traffic analysis result sent by a network device, wherein the switching chip of the network device sends the target traffic message currently carried by the target port queue to the CPU of the network device when determining that the utilization rate of the target port queue buffer reaches a preset high threshold; the CPU analyzes the target traffic message to obtain the target traffic analysis result;
[0101] The determining unit 41 is configured to determine a cause of congestion of the target port queue based on the target traffic analysis result and the historical traffic analysis result of the target port queue.
[0102] Optionally, the target traffic analysis result includes traffic component information of the target traffic, and the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow;
[0103] When determining a cause of congestion of the target port queue based on the target traffic analysis result and the historical traffic analysis result of the target port queue, the determining unit 41 is specifically configured to:
[0104] Based on the source IP, destination IP, protocol type, and packet size of each target flow, as well as the source IP, destination IP, protocol type, and packet size of each flow included in the historical traffic analysis results, determine whether there is a newly added first flow in the traffic carried by the target port queue, and determine whether there is a second flow with an increase in traffic bandwidth greater than a set threshold;
[0105] If it is determined that the first flow and / or the second flow exist, the first flow and / or the second flow are determined as the cause of congestion of the target port queue.
[0106] The above units may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a unit is implemented by scheduling program code through a processing element, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these units may be integrated together and implemented in the form of a system-on-a-chip (SOC).
[0107] Furthermore, the congestion handling device provided in the embodiment of the present application, from the hardware level, the hardware architecture diagram of the congestion handling device can be found in Figure 5 As shown, the congestion processing device may include: a memory 50 and a processor 51,
[0108] The memory 50 is used to store program instructions. The processor 51 calls the program instructions stored in the memory 50 and executes the above method embodiment applied to the network device according to the obtained program instructions. The specific implementation method and technical effect are similar and will not be repeated here.
[0109] Optionally, the present application also provides a network device, comprising at least one processing element (or chip) for executing the above-mentioned method embodiment applied to the network device.
[0110] Optionally, the present application also provides a program product, such as a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable the computer to execute the above-mentioned method embodiment applied to the network device.
[0111] Furthermore, the congestion handling device provided in the embodiment of the present application, from the hardware level, the hardware architecture diagram of the congestion handling device can be found in Figure 6 As shown, the congestion processing device may include: a memory 60 and a processor 61,
[0112] The memory 60 is used to store program instructions. The processor 61 calls the program instructions stored in the memory 60 and executes the above-mentioned method embodiment applied to the analyzer according to the obtained program instructions. The specific implementation method and technical effect are similar and will not be repeated here.
[0113] Optionally, the present application also provides an analyzer, comprising at least one processing element (or chip) for executing the above-mentioned method embodiment applied to the analyzer.
[0114] Optionally, the present application also provides a program product, such as a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable the computer to execute the above-mentioned method embodiment applied to the analyzer.
[0115] Here, the machine-readable storage medium can be any electronic, magnetic, optical or other physical storage device that can contain or store information, such as executable instructions, data, etc. For example, the machine-readable storage medium can be: RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard disk drive), solid state drive, any type of storage disk (such as CD, DVD, etc.), or similar storage media, or a combination thereof.
[0116] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer, which may be in the form of a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email transceiver, game console, tablet computer, wearable device, or any combination of these devices.
[0117] For the convenience of description, the above devices are described as being divided into various units according to their functions. Of course, when implementing this application, the functions of each unit can be implemented in the same or multiple software and / or hardware.
[0118] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the embodiments of the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0119] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0120] Furthermore, these computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0121] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.
[0122] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.
Claims
1. A congestion handling method, characterized in that: Applied to a network device; the method comprises: The switching chip of the network device monitors the utilization rate of each port queue buffer; If the switching chip determines that the target port queue buffer utilization reaches a preset high threshold, the target traffic message currently carried by the target port queue is sent to the CPU of the network device; The CPU analyzes the target traffic message to obtain the target traffic analysis result; The CPU sends the target traffic analysis result to the analyzer, so that the analyzer determines a cause of congestion of the target port queue based on the target traffic analysis result and a historical traffic analysis result of the target port queue.
2. The method according to claim 1, wherein If the switching chip determines that the utilization rate of the target port queue buffer reaches a preset high threshold, the step of sending the target traffic message currently carried by the target port queue to the CPU includes: If the switching chip determines that the utilization rate of the target port queue buffer reaches a preset high threshold, the port state of the target port is marked as a burst state; The switching chip triggers a congestion event report and starts a message copy function to copy the target traffic message carried by the target port queue and send it to the CPU.
3. The method according to claim 1 or 2, wherein: The CPU analyzes the target traffic message to obtain the target traffic analysis result, including: The CPU analyzes the traffic component information of the target traffic based on the target traffic message, wherein the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow.
4. A congestion handling method, characterized in that: Applied to an analyzer, the method comprises: receiving a target traffic analysis result sent by a network device, wherein, when determining that a utilization rate of a target port queue buffer reaches a preset high threshold, the switching chip of the network device sends the target traffic message currently carried by the target port queue to a CPU of the network device; the CPU analyzes the target traffic message to obtain the target traffic analysis result; Based on the historical traffic analysis result of the target port queue and the target traffic analysis result, a cause of congestion of the target port queue is determined.
5. The method according to claim 4, wherein The target traffic analysis result includes traffic component information of the target traffic, and the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow; The step of determining a cause of congestion of the target port queue based on the target traffic analysis result and a historical traffic analysis result of the target port queue includes: Based on the source IP, destination IP, protocol type, and packet size of each target flow, as well as the source IP, destination IP, protocol type, and packet size of each flow included in the historical traffic analysis results, determine whether there is a newly added first flow in the traffic carried by the target port queue, and determine whether there is a second flow with an increase in traffic bandwidth greater than a set threshold; If it is determined that the first flow and / or the second flow exist, the first flow and / or the second flow are determined as the cause of congestion of the target port queue.
6. A network device, characterized in that: The network device includes a switching chip and a processor: The switch chip is used to monitor the utilization of each port queue buffer; If the switching chip determines that the target port queue buffer utilization reaches a preset high threshold, the target traffic message currently carried by the target port queue is sent to the processor of the network device; The processor is configured to analyze the target traffic message to obtain the target traffic analysis result; The processor is further configured to send the target traffic analysis result to an analyzer, so that the analyzer determines a cause of congestion of the target port queue based on the target traffic analysis result and a historical traffic analysis result of the target port queue.
7. The network device according to claim 6, wherein: If the switching chip determines that the utilization rate of the target port queue buffer reaches a preset high threshold, then when sending the target traffic message currently carried by the target port queue to the processor, the switching chip is specifically configured to: If it is determined that the target port queue buffer utilization reaches a preset high threshold, the port state of the target port is marked as a burst state; The congestion event report is triggered, and the message copy function is started to copy the target traffic message carried by the target port queue and send it to the processor.
8. The network device according to claim 6 or 7, wherein: When analyzing the target traffic message and obtaining the target traffic analysis result, the processor is specifically configured to: Based on the target traffic message, the traffic component information of the target traffic is analyzed, wherein the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow.
9. A congestion handling device, characterized in that: Applied to an analyzer, the device comprises: a receiving unit, configured to receive a target traffic analysis result sent by a network device, wherein the switching chip of the network device sends the target traffic message currently carried by the target port queue to a CPU of the network device when determining that the utilization rate of the target port queue buffer reaches a preset high threshold; the CPU analyzes the target traffic message to obtain the target traffic analysis result; The determining unit is configured to determine a cause of congestion of the target port queue based on a historical traffic analysis result of the target port queue and the target traffic analysis result.
10. The device according to claim 9, characterized in that The target traffic analysis result includes traffic component information of the target traffic, and the traffic component information includes the source IP, destination IP, protocol type and data packet size of each target flow; When determining a cause of congestion of the target port queue based on the target traffic analysis result and a historical traffic analysis result of the target port queue, the determining unit is specifically configured to: Based on the source IP, destination IP, protocol type, and packet size of each target flow, as well as the source IP, destination IP, protocol type, and packet size of each flow included in the historical traffic analysis results, determine whether there is a newly added first flow in the traffic carried by the target port queue, and determine whether there is a second flow with an increase in traffic bandwidth greater than a set threshold; If it is determined that the first flow and / or the second flow exist, the first flow and / or the second flow are determined as the cause of congestion of the target port queue.
11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable the computer to execute the steps of the method according to any one of claims 1 to 3 or any one of claims 4 to 5.