CN115834484A - Method, device and equipment for detecting transient congestion on chip and storage medium - Google Patents

Abstract

The application discloses a method, a device, equipment and a storage medium for detecting transient congestion on a chip, wherein the method comprises the following steps: acquiring the resource occupancy rate of a chip port queue; judging whether the port queue meets a preset transient congestion capturing condition or not according to the resource occupancy rate, and triggering a preset capturing engine to enter a transient congestion capturing state to capture transient congestion event information of the port queue when the transient congestion capturing condition is met; storing the captured transient congestion event information. According to the transient congestion detection method provided by the application, transient congestion events can be captured and recorded in the extremely short duration of transient congestion, and the scheme can be applied to the extremely fine detection granularity of the port queue, so that the subsequent analysis is facilitated, and the influence of transient congestion flow on the network and the application performance is reduced.

Description

On-chip transient congestion detection method, device, equipment and storage medium

technical field

The invention relates to the technical field of network communication, in particular to a method, device, equipment and storage medium for instantaneous congestion detection on a chip.

Background technique

Mircoburst refers to the phenomenon of bursty transmission of a large number of network data packets in a chip port within a very short time range. For extremely short durations, the instantaneous rate may far exceed the average port rate. This transient congestion can have a significant negative impact on network and application performance, exacerbating latency, jitter, and packet loss in the network.

At present, the monitoring methods commonly used in the network have a long time dimension, and it is very difficult to capture and analyze instantaneous congestion. Moreover, the reporting of frequent transient congestion events also poses a great challenge to on-chip resource consumption and CPU access bandwidth.

Contents of the invention

Embodiments of the present application provide an on-chip transient congestion detection method, device, device, and storage medium. In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is presented below. This summary is not an overview, nor is it intended to identify key/critical elements or delineate the scope of these embodiments. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is presented later.

In the first aspect, the embodiment of the present application provides an on-chip transient congestion detection method, including:

Obtain the resource occupancy rate of the chip port queue;

Judging whether the port queue satisfies the preset transient congestion capture condition according to the resource occupancy rate, when the transient congestion capture condition is satisfied, triggering the preset capture engine to enter the transient congestion capture state, and capturing the transient congestion event information of the port queue;

The captured transient congestion event information is stored.

In an optional embodiment, judging whether the port queue satisfies a preset instantaneous congestion capture condition according to the resource occupancy rate includes:

Determine whether the resource occupancy rate of the port queue is greater than or equal to the preset upper threshold;

Determine whether the enable switch of the port queue is on;

When the resource occupancy rate of the port queue is greater than or equal to the upper threshold and the enabling switch is in an open state, it is determined that the port queue satisfies a preset instantaneous congestion capture condition.

In an optional embodiment, before judging whether the resource occupancy rate of the port queue is greater than or equal to a preset upper threshold, it also includes:

The upper threshold is dynamically adjusted according to the port queue rate and the current resource occupancy rate of the port queue.

In an optional embodiment, after the preset capture engine is triggered to enter the transient congestion capture state, it further includes:

Determine whether the current port queue resource occupancy rate is less than the preset lower threshold;

When the port queue resource occupancy rate is less than the preset threshold lower limit, trigger the preset capture engine to exit the transient congestion capture state;

Judging whether the instantaneous congestion capture time of the current port queue is greater than the preset capture time threshold;

When the transient congestion capture time of the port queue is greater than the preset capture time threshold, the preset capture engine is triggered to exit the transient congestion capture state.

In an optional embodiment, capturing the transient congestion event information of the port queue includes:

The capture start time of the capture port queue, the capture port queue number, the capture end time, the peak value of transient congestion, and the total number of times the port queue enters the capture state.

In an optional embodiment, storing the captured transient congestion event information includes:

Storing the transient congestion event information into the cache module of the capture engine;

Merge the instantaneous congestion events within the preset merging time window stored by the cache module to obtain the merged instantaneous congestion event information;

The combined transient congestion event information is stored in a storage module and/or in a remote sensing packager.

In an optional embodiment, the instantaneous congestion events stored in the cache module within the preset merging time window are merged to obtain the merged instantaneous congestion event information, including:

Obtain the maximum resource occupancy rate of all transient congestion events within the preset combined time window as the maximum resource occupancy rate of the combined transient congestion event;

Obtain the sum of the times of all transient congestion events within the preset combined time window as the combined number of transient congestion events;

Obtain the minimum value of the capture start time of all transient congestion events within the preset combined time window as the capture start time of the combined transient congestion event;

Obtain the maximum value of the capture end time of all transient congestion events within the preset merge time window, and use it as the capture end time of the merged transient congestion event.

In the second aspect, the embodiment of the present application provides an on-chip transient congestion detection device, including:

An acquisition module, configured to acquire the resource occupancy rate of the chip port queue;

The capture module is used to judge whether the port queue satisfies the preset transient congestion capture condition according to the resource occupancy rate, and when the transient congestion capture condition is met, trigger the preset capture engine to enter the transient congestion capture state, and capture the port queue Transient congestion event information;

A storage module, configured to store the captured transient congestion event information.

In a third aspect, an embodiment of the present application provides an electronic device, including a processor and a memory storing program instructions, and the processor is configured to, when executing the program instructions, execute the on-chip Transient congestion detection method.

In a fourth aspect, the embodiment of the present application provides a computer-readable medium on which computer-readable instructions are stored, and the computer-readable instructions are executed by a processor to implement the method for detecting transient congestion on a chip provided by the above-mentioned embodiment.

The technical solutions provided by the embodiments of the present application may include the following beneficial effects:

The transient congestion detection method provided by the embodiment of the present application judges whether the transient congestion capture condition is met according to the resource occupation of the port. When the capture condition is satisfied, the preset capture engine is controlled to perform the transient congestion capture state, and the transient congestion event on the chip is detected. information is captured and stored. It can capture and record transient congestion events in a very short duration of transient congestion, which is convenient for subsequent analysis and reduces the impact of transient congestion traffic on network and application performance, and the capture granularity can be accurate to the port queue level.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention.

Description of drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description serve to explain the principles of the invention.

FIG. 1 is a schematic flowchart of a method for detecting transient congestion on a chip according to an exemplary embodiment;

Fig. 2 is a schematic diagram of an on-chip transient congestion detection method according to an exemplary embodiment;

Fig. 3 is a schematic diagram of an on-chip transient congestion detection method according to an exemplary embodiment;

Fig. 4 is a schematic diagram showing detection conditions of transient congestion traffic according to an exemplary embodiment;

Fig. 5 is a schematic structural diagram of an on-chip transient congestion detection device according to an exemplary embodiment;

Fig. 6 is a schematic structural diagram of an electronic device according to an exemplary embodiment;

Fig. 7 is a schematic diagram of a computer storage medium according to an exemplary embodiment.

Detailed ways

The following description and drawings illustrate specific embodiments of the invention sufficiently to enable those skilled in the art to practice them.

It should be clear that the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

When the following description refers to the accompanying drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the present invention. Rather, they are merely examples of systems and methods consistent with aspects of the invention as recited in the appended claims.

The on-chip transient congestion detection method provided by the embodiment of the present application will be described in detail below with reference to the accompanying drawings. Referring to Fig. 1, the method specifically includes the following steps.

S101 acquires the resource occupancy rate of the chip port queue.

In one embodiment, the transient congestion traffic on the switch chip is detected, and the detection object is the resource occupation of ports or port queues in the chip. Those skilled in the art can detect transient port congestion events according to actual needs, or detect the resource occupancy of each queue included in the port in a more fine-grained manner.

In a possible implementation, the source of capture is divided into event capture and background scan capture. Event capture refers to the change of on-chip resources caused by data packets entering and exiting ports or port queues. Therefore, when data packets enter and exit, the collected data Port resource occupancy rate or port queue resource occupancy rate when packets are in and out. The background scan is used as a supplement to the event capture, and scans on-chip resources when no event capture occurs.

S102 judges whether the port queue satisfies the preset transient congestion capture condition according to the resource occupancy rate, and when the transient congestion capture condition is satisfied, triggers the preset capture engine to enter the transient congestion capture state, and captures the transient congestion event information of the port queue.

In an exemplary embodiment, multiple capture engines are pre-configured on the switch chip, and each capture engine can support multiple ports and queues. Among them, the number of capture engines can be flexibly increased or decreased according to the number of entry and exit planes in the chip, and the actual needs of the required detection ports or queues. For example, a capture engine can be configured for each port, or more finely-grained, a capture engine can be configured for each queue, which can be set by those skilled in the art according to actual needs.

Further, according to the obtained port queue resource occupation, it is judged whether to enter the instantaneous congestion capture. Including: judging whether the resource occupancy rate of the port queue is greater than or equal to the preset upper threshold, judging whether the enable switch of the port queue is on, when the resource occupancy rate of the port queue is greater than or equal to the upper threshold, and the enable switch is on , to determine that the port queue satisfies the preset instantaneous congestion capture condition.

Wherein, in the solution provided by the embodiment of the present application, the upper threshold can be dynamically adjusted according to the port queue rate and the current resource occupancy rate of the port queue. A dynamic threshold is built into the detection and judgment. According to the port queue rate and the current port queue status, the upper limit of the threshold is dynamically adjusted. When the actual resource usage is high, the dynamic threshold is appropriately increased, thereby reducing the number of required captures and saving storage resources and chips. power consumption. For port queues with different rates, configure different port queue rate coefficients to reduce the number of static threshold configuration items, and at the same time ensure that port queues with different rates adopt different upper thresholds.

In a possible implementation, the upper threshold can be determined according to the following formula:

Upper threshold = static threshold + port queue rate factor * (actual resource usage - static threshold);

Among them, the port queue rate coefficient and the static threshold are software configuration items, which can be configured by those skilled in the art according to the actual situation, and are not specifically limited in this embodiment of the application.

Further, after it is determined that the port queue satisfies the preset transient congestion capture condition, a preset capture engine is triggered to enter the transient congestion capture state, and the transient congestion event information of the port is captured.

In an optional embodiment, the capture engine captures the capture start time of the port or the port queue, the capture port or port queue number, the capture end time, the transient congestion peak value, and the total number of times the port or the port queue enters the capture state. A capture counter is set in the capture state, which can count the total number of times the port or port queue enters the capture state.

After entering the capture state, it also includes judging whether the transient congestion event is over. In an exemplary embodiment, it also includes setting a lower threshold, judging whether the current port resource occupancy rate is lower than the preset lower threshold; when the port resource occupancy rate is lower than the preset lower threshold, triggering the preset capture engine to exit transient congestion capture state. In addition, it also includes a software-configurable capture exit maximum time. When the capture state lasts longer than the preset capture time threshold, the capture engine will exit the capture state. This function ensures that when a port is congested for a long time, other ports can still are captured in the same engine. Wherein, the specific value of the preset capture time threshold can be set according to the actual situation.

Wherein, the lower threshold can be a fixed value, such as setting the lower threshold according to a static threshold, or it can be a dynamic value, which is automatically adjusted following the upper threshold according to the port rate and actual resource occupation. For example, the lower limit of the threshold is determined according to the following formula: lower limit of the threshold = upper limit of the threshold - fluctuation coefficient, and the fluctuation coefficient can be set according to the actual situation. Through different upper and lower thresholds, it can effectively alleviate the miscapture when the resource jitters back and forth near the single threshold.

According to this step, the transient congestion on the switch chip can be detected, and information such as the start time information, end time information, burst peak value, port queue number entering the capture state and the like of the transient congestion event can be captured.

Each transient congestion capture engine can provide the CPU with real-time access status information: whether it is currently under transient congestion monitoring; the port number and queue number of the currently monitored transient congestion; the maximum resource occupation of the currently detected transient congestion port number ; The total number of times a transient congestion detection has occurred.

S103 stores the captured transient congestion event information.

In a possible implementation, in each transient congestion capture engine, a cache module is set to temporarily store the captured data before the transient congestion event information is stored in the shared memory, and there is also a built-in merging mechanism in the cache , within a time window configurable by the software, the capture events of the same port or the capture events of the same port queue are merged, thereby reducing storage space and bandwidth required for software access.

Specifically, the transient congestion event information is stored in the cache module of the capture engine, and the transient congestion events within the preset combining time window stored in the cache module are merged to obtain the merged transient congestion event information.

In one embodiment, the instantaneous congestion events of the same port or the same port queue within the preset merging time window are acquired and merged. The maximum resource occupancy rate of all transient congestion events within the preset combined time window is used as the maximum resource occupancy rate of the combined transient congestion events; the sum of the times of all transient congestion events within the preset combined time window is used as the combined The number of transient congestion events after the combination; the minimum value of the capture start time of all transient congestion events in the preset merging time window is used as the capture start time of the combined transient congestion event; all the transient congestion events in the preset merging time window The maximum capture end time of is used as the capture end time of the merged transient congestion event.

As shown in Figure 4, it is the obtained port resource occupancy. When the resource occupancy is greater than the upper threshold, the capture is triggered, and when it falls to the lower threshold, the capture ends. And a merge time window is set, and the instantaneous congestion events within the set time period are merged.

Furthermore, in the transient congestion detection solution provided by this application, in order to improve storage efficiency and save chip resources, a fixed-depth shared storage space is allocated according to actual needs to store the record data generated by all capture engines. The shared memory space requires only a single-port static random access memory (SRAM). The capture engine can store the merged transient congestion event information into a shared storage module for the CPU to read transient congestion traffic detection data at any time. Compared with real-time reporting, it saves a lot of bandwidth occupation of the CPU access path.

In an optional embodiment, the merged transient congestion event information can also be stored in the remote sensing packetizer, and the transient congestion information can be inserted into the LNS (local name space) domain segment of the IFA (Inband Flow Analyzer) packet , so as to extend the function of sending the instantaneous congestion information of the local exchange in the form of network remote sensing data packets. Instantaneous congestion capture information can be obtained without using the CPU.

In order to facilitate the understanding of the on-chip transient congestion detection method provided by the embodiment of the present application, the following description will be made with reference to FIG. 2 .

As shown in FIG. 2 , the transient congestion detection method provided by the embodiment of the present application includes several steps of background scanning, event change scanning, detection and judgment, state capture, buffering, and storage.

First of all, when the data packet enters and exits the port, the change of the on-chip resource and the occupancy of the on-chip resource when there is no data packet enters and exits. Then, detection and judgment are carried out to judge whether to enter the instantaneous congestion capture. Including: judging whether the resource occupancy rate of the port is greater than or equal to the preset upper threshold, judging whether the enable switch of the port is on, when the resource occupancy rate of the port is greater than or equal to the upper threshold and the enable switch is on, determine the port Satisfy the preset transient congestion capture conditions. After confirming that the capture conditions are satisfied, the capture engine is triggered to capture the state, and record the start time information, end time information, burst peak value, port queue number entering the capture state and other information of the transient congestion event.

Further, the transient congestion event information is stored in the cache module of the capture engine, and the transient congestion events within the preset combining time window stored in the cache module are merged to obtain the merged transient congestion event information. The capture engine can store the merged transient congestion event information into a shared storage module for the CPU to read transient congestion flow detection data. The merged instantaneous congestion event information can also be stored in the remote sensing packetizer, thereby expanding the function of sending the instantaneous congestion information of the local exchange in the form of network remote sensing data packets.

In order to facilitate the understanding of the on-chip transient congestion detection method provided by the embodiment of the present application, it will be described below with reference to FIG. 3 .

As shown in FIG. 3 , the transient congestion detection method provided by the embodiment of the present application includes: firstly, acquiring on-chip resource changes when data packets enter and exit ports, and on-chip resource occupancy when no data packets enter and exit. Then, detection and judgment are carried out to judge whether to enter the instantaneous congestion capture. Including: judging whether the resource occupancy rate of the port is greater than or equal to the preset upper threshold, judging whether the enable switch of the port is on, when the resource occupancy rate of the port is greater than or equal to the upper threshold and the enable switch is on, determine the port Satisfy the preset transient congestion capture conditions.

Further, it is judged whether the capture duration is overdue and whether the resource occupancy rate is lower than the lower threshold, and if the capture duration is overtime or the resource occupancy rate is lower than the lower threshold, exit the capture state and save the capture information.

According to the transient congestion detection method provided by the embodiment of the present application, it is judged whether the transient congestion capture condition is satisfied according to the resource occupation of the port. Event information is captured and stored. Capable of capturing and recording transient congestion events in a very short duration of transient congestion, which facilitates subsequent analysis and reduces the impact of transient congestion traffic on network and application performance.

Moreover, the transient congestion detection scheme in the switch chip proposed in this application provides highly flexible and configurable capturing and recording based on emergencies. While ensuring fine-grained detection, the capture window and frequency can be adaptively adjusted to save chip resources occupied by the detection process and CPU access bandwidth. This solution is applicable to both outgoing ports and queues, as well as incoming ports and queues. It does not depend on the main data path inside the chip.

The embodiment of the present application also provides an on-chip transient congestion detection device, which is used to implement the on-chip transient congestion detection method of the above embodiment, as shown in FIG. 5 , the device includes:

An acquisition module 501, configured to acquire the resource occupancy rate of the chip port queue;

The capture module 502 is used to judge whether the port queue meets the preset transient congestion capture condition according to the resource occupancy rate, and when the transient congestion capture condition is met, trigger the preset capture engine to enter the transient congestion capture state, and capture the transient congestion event of the port queue information;

The storage module 503 is configured to store captured instantaneous congestion event information.

It should be noted that when the device for detecting transient congestion on a chip provided by the above-mentioned embodiments executes the method for detecting transient congestion on a chip, it only uses the division of the above-mentioned functional modules for illustration. In practical applications, the above-mentioned Function allocation is accomplished by different functional modules, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the on-chip transient congestion detection device and the on-chip transient congestion detection method embodiment provided by the above embodiments belong to the same concept, and the implementation process thereof is detailed in the method embodiment, and will not be repeated here.

An embodiment of the present application further provides an electronic device corresponding to the method for detecting transient congestion on a chip provided in the foregoing embodiments, so as to implement the method for detecting transient congestion on a chip.

Please refer to FIG. 6 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. As shown in Figure 6, the electronic equipment includes: processor 600, memory 601, bus 602 and communication interface 603, processor 600, communication interface 603 and memory 601 are connected by bus 602; A running computer program, when the processor 600 runs the computer program, executes the on-chip transient congestion detection method provided in any one of the foregoing embodiments of the present application.

Wherein, the memory 601 may include a high-speed random access memory (RAM: Random Access Memory), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 603 (which may be wired or wireless), and the Internet, wide area network, local network, metropolitan area network, etc. can be used.

The bus 602 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus can be divided into address bus, data bus, control bus and so on. Wherein, the memory 601 is used to store the program, and the processor 600 executes the program after receiving the execution instruction. The on-chip transient congestion detection method disclosed in any implementation manner of the above-mentioned embodiments of the present application can be applied to the processor 600, or by The processor 600 implements.

The processor 600 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method may be implemented by an integrated logic circuit of hardware in the processor 600 or an instruction in the form of software. The above-mentioned processor 600 can be a general-purpose processor, including a central processing unit (Central Processing Unit, referred to as CPU), a network processor (Network Processor, referred to as NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), off-the-shelf programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. Various methods, steps, and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like. The steps of the method disclosed in the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register. The storage medium is located in the memory 601, and the processor 600 reads the information in the memory 601, and completes the steps of the above method in combination with its hardware.

The electronic device provided in the embodiment of the present application is based on the same inventive concept as the method for detecting transient congestion on a chip provided in the embodiment of the present application, and has the same beneficial effect as the method adopted, operated or implemented.

The embodiment of the present application also provides a computer-readable storage medium corresponding to the on-chip transient congestion detection method provided in the foregoing embodiments. Please refer to FIG. A computer program (that is, a program product) is stored, and when the computer program is run by the processor, it will execute the on-chip transient congestion detection method provided by any of the foregoing embodiments.

It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access Memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media will not be repeated here.

The computer-readable storage medium provided by the above-mentioned embodiments of the present application is based on the same inventive concept as the on-chip transient congestion detection method provided by the embodiments of the present application, and has the same method adopted, run or implemented by the stored application program. Beneficial effect.

The technical features of the above embodiments can be combined arbitrarily. To make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be It is considered to be within the range described in this specification.

The above examples only express several implementations of the present invention, and the description thereof is relatively specific and detailed, but should not be construed as limiting the patent scope of the present invention. It should be pointed out that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention should be based on the appended claims.

Claims (10)

1. A transient congestion detection method on a chip is characterized by comprising the following steps:

acquiring the resource occupancy rate of a chip port queue;

judging whether the port queue meets a preset transient congestion capturing condition or not according to the resource occupancy rate, and triggering a preset capturing engine to enter a transient congestion capturing state to capture transient congestion event information of the port queue when the transient congestion capturing condition is met;

storing the captured transient congestion event information.

2. The method of claim 1, wherein determining whether a port queue meets a preset transient congestion capture condition based on the resource occupancy comprises:

judging whether the resource occupancy rate of the port queue is greater than or equal to a preset threshold upper limit;

judging whether an enabling switch of the port queue is in an open state or not;

and when the resource occupancy rate of the port queue is greater than or equal to the upper limit of the threshold value and the enabling switch is in an open state, determining that the port queue meets a preset transient congestion capturing condition.

3. The method of claim 2, wherein before determining whether the resource occupancy of the port queue is greater than or equal to a preset upper threshold, the method further comprises:

and dynamically adjusting the upper limit of the threshold value according to the rate of the port queue and the current resource occupancy rate of the port queue.

4. The method of claim 1, wherein after triggering the predetermined trapping engine to enter a transient congestion trapping state, further comprising:

judging whether the resource occupancy rate of the current port queue is smaller than a preset threshold lower limit;

when the resource occupancy rate of the port queue is lower than the preset lower threshold, triggering a preset capture engine to exit from a transient congestion capture state;

judging whether transient congestion capture time of the current port queue is greater than a preset capture time threshold value or not;

and when the transient congestion capture time of the port queue is greater than a preset capture time threshold, triggering a preset capture engine to exit from a transient congestion capture state.

5. The method of claim 1, wherein capturing transient congestion event information for a port queue comprises:

the acquisition start time of the acquisition port queue, the acquisition port queue number, the acquisition end time, the instantaneous congestion spike value and the total number of times the port queue enters the acquisition state.

6. The method of claim 1 wherein storing the captured transient congestion event information comprises:

storing the transient congestion event information into a cache module of a capture engine;

merging the transient congestion events stored in the cache module within a preset merging time window to obtain merged transient congestion event information;

and storing the merged transient congestion event information into a storage module and/or a remote sensing group packaging device.

7. The method of claim 6, wherein merging the transient congestion events stored in the preset merging time window by the cache module to obtain merged transient congestion event information comprises:

acquiring the maximum resource occupancy rate of all transient congestion events in a preset merging time window, and taking the maximum resource occupancy rate as the maximum resource occupancy rate of the merged transient congestion events;

acquiring the sum of the times of all transient congestion events in a preset merging time window as the number of merged transient congestion events;

acquiring the minimum value of the capture start time of all transient congestion events in a preset merging time window, and taking the minimum value as the capture start time of the merged transient congestion events;

and acquiring the maximum value of the capturing end time of all transient congestion events in the preset merging time window as the capturing end time of the merged transient congestion events.

8. An apparatus for detecting transient congestion on a chip, comprising:

the acquisition module is used for acquiring the resource occupancy rate of the chip port queue;

the capturing module is used for judging whether the port queue meets a preset transient congestion capturing condition according to the resource occupancy rate, and triggering a preset capturing engine to enter a transient congestion capturing state to capture transient congestion event information of the port queue when the transient congestion capturing condition is met;

and the storage module is used for storing the captured transient congestion event information.

9. An electronic device comprising a processor and a memory storing program instructions, the processor being configured to perform the on-chip transient congestion detection method of any of claims 1 to 7 when executing the program instructions.

10. A computer readable medium having computer readable instructions stored thereon which are executable by a processor to implement a method of transient congestion detection on a chip as claimed in any one of claims 1 to 7.

Images