CN113392607B - A method for determining configuration parameters and related equipment - Google Patents

Abstract

This application discloses a method for determining configuration parameters and related equipment, including: determining forwarding service simulation information, which includes input traffic information and data path behavior; and separately determining the forwarding service simulation information based on multiple sets of first configuration parameters. A simulation test is performed to obtain multiple sets of first simulation test results; and based on the multiple sets of first simulation test results, a target first configuration parameter corresponding to the target first simulation test result is determined from the multiple sets of first configuration parameters. Through simulation testing, the efficiency of determining configuration parameters can be effectively improved. Determining configuration parameters does not require technicians to have high technical analysis capabilities and rich debugging experience. Moreover, the determined configuration parameters can meet the needs of multiple business overlays. the required data path performance.

Description

A method for determining configuration parameters and related equipment

Technical field

This application relates to the technical field of simulation testing, and in particular to a method for determining configuration parameters and related equipment.

Background technique

Data path refers to the data transmission path formed by connecting various subsystems through a data bus in a digital system. Under normal circumstances, there may be many modules on the data path to control the traffic on the data path accordingly, such as physical address aggregation (MAC aggregation, MAG) module, switching interface control (fabric interface control, FIC) module, traffic management ( traffic management (TM) module and packet reassemblyblocks (PRB) module. After corresponding parameter configuration, these modules can control the priority and rate of the output traffic of the data path. Therefore, for the parameter configuration of each module, , which has an important impact on the performance of the data path.

At present, the data path of the product (such as a chip, a single board, etc.) is usually tested on the test equipment, and the technical personnel gradually adjust the configured parameters of each module on the product data path based on the test results, thereby passing the technical personnel After multiple on-board tests and configuration parameter adjustments, the configuration parameters that can enable the product to meet the expected test requirements are obtained. However, this method of determining configuration parameters requires high skills and difficulty for technical personnel, and the efficiency of determining configuration parameters is also low.

Contents of the invention

Embodiments of the present application provide a method for determining configuration parameters and related equipment, so as to reduce the skill requirements for technical personnel in the process of determining configuration parameters and improve the efficiency of determining configuration parameters.

In the first aspect, embodiments of the present application provide a method for determining configuration parameters. Specifically, forwarding service simulation information is first determined. The forwarding service simulation information includes input traffic information and data path behavior, where the input traffic information is In order to simulate input traffic corresponding to multiple forwarding services, the data path behavior is used to simulate a data path that transmits the input traffic; then, the forwarding services can be configured based on multiple sets of first configuration parameters of the data path. The simulation information is subjected to simulation testing to obtain multiple sets of first simulation test results. The multiple sets of first configuration parameters are different configuration parameters. Each set of first simulation test results in the multiple sets of first simulation test results is used to simulate the The output traffic after the input traffic corresponding to the multiple forwarding services passes through the data path corresponding to a set of first configuration parameters in the multiple sets of first configuration parameters; in this way, according to the multiple sets of first simulation test results, The target first configuration parameter corresponding to the target first simulation test result can be determined from the plurality of sets of first configuration parameters.

In this embodiment, simulation testing is used to determine the configuration parameters corresponding to each module on the data path. This can effectively improve the accuracy of determining the configuration parameters compared to repeated trial and error in the on-board test environment. Efficiency also saves the overhead required for board testing. Moreover, the simulation platform can traverse multiple sets of configuration parameters and provide simulation test results corresponding to each set of configuration parameters. It does not require technicians to have high technical analysis capabilities and rich debugging experience, and can also determine from multiple sets of configuration parameters. The required configuration parameters can be identified, thereby reducing the skill requirements for technical personnel. At the same time, the determined configuration parameters can meet the data path performance required when multiple forwarding services are superimposed. This can avoid the data path performance of the product only meeting the data path performance in one business scenario but not another business. Data path performance under the scenario.

As an example, the simulation platform can be an ESL simulation platform, or other simulation platforms.

In conjunction with the first aspect, in a first possible implementation manner of the first aspect, the input traffic information includes any one or more of a service type, a port template, a rate template, a priority template, and a packet length template, The data path behavior includes port information and/or service forwarding paths. In this embodiment, one or more templates can be used to simulate the input traffic of the forwarding service, and the data path behavior port information and/or the service forwarding path can be used to simulate the data path of the input traffic, thereby simulating the forwarding service.

With reference to the first implementation manner of the first aspect, in a second possible implementation manner of the second aspect, the service type includes unicast and/or multicast; the port template includes spontaneous self-collection, half-offset , symmetrical offset, random offset, any one or more of leaf full coverage, leaf front half coverage, leaf back half coverage, leaf interval half coverage and leaf random coverage; the rate template includes burst on, burst Any one or more of unopened, full flow, half flow and custom percentage traffic; the priority template includes one or more traffic transmission priorities, and when the priority template includes multiple traffic transmission In priority, the ratio between the multiple traffic transmission priorities is a preset ratio and/or a custom ratio; the packet length template includes a custom packet length, and the packet length randomly jumps within the preset packet length interval. , the packet length increases and jumps within the preset packet length interval, and any one or more of a variety of packet length mixtures; the port information includes any one of port number, physical address aggregation MAG source, and port rate, or Multiple; the service forwarding path is used to describe the forwarding path and/or forwarding delay of the service.

With reference to any one of the first to second implementations of the first aspect, in a third possible implementation of the first aspect, the configuration parameters of the target module on the data path are the same as the The members in the target class corresponding to the target module are assigned values, and different target modules on the data path correspond to different target classes. In this implementation, the modules on the data path can be simulated through defined classes, and the configuration parameters of each module on the data path are assigned values to the members in the corresponding classes. In practical applications, the data path or the equipment (such as single boards, chips, etc.) carrying the data path can be abstractly defined as a large class. Each module on the data path is an internal class, and each module on the data path performs the parameter configuration process. , that is, the process of assigning values to members in the corresponding classes for the simulation platform, so that the simulation test of forwarding business simulation information can be realized.

With reference to any one of the first to third implementations of the first aspect, in a fourth possible implementation of the first aspect, the method further includes: according to the plurality of sets of first configurations Parameters and the plurality of sets of first simulation test results, a machine learning model is obtained by training. The input of the machine learning model is each group of first configuration parameters in the plurality of groups of first configuration parameters. The output of the machine learning model is the first simulation test result corresponding to each set of first configuration parameters. In this implementation, after using the simulation platform to obtain simulation test results corresponding to multiple sets of configuration parameters, the configuration parameters and their corresponding simulation test results can be used to train a machine learning model. The machine learning model can be obtained based on the input configuration parameter output. Corresponding simulation test results. In this way, when you need to obtain the simulation test results corresponding to a certain configuration parameter, you can obtain the simulation test results without using simulation. Instead, you can directly use the machine learning model output to obtain the required simulation test results. .

With reference to the fourth implementation manner of the first aspect, in a fifth possible implementation manner of the first aspect, the method further includes: determining multiple sets of different second configuration parameters; Configuration parameters are respectively input into the machine learning model to obtain multiple sets of second simulation test results, and the multiple sets of second simulation test results are outputs of the machine learning model; the target is determined from the multiple sets of second simulation test results. The second simulation test result, the target second simulation test result is consistent with the expected output traffic, and the expected output traffic is the output traffic after the input traffic of the target forwarding service passes through the data path; determine the target second simulation test The second configuration parameter corresponding to the result. In this implementation, a machine learning model can be used to determine from a set of configuration parameters configuration parameters that meet the data path performance requirements in the target forwarding business scenario, so that there is no need to determine the configuration through simulation testing or board testing. parameters, effectively improving the efficiency of determining configuration parameters.

In combination with the fourth implementation manner or the fifth implementation manner of the first aspect, in a sixth possible implementation manner of the first aspect, the machine learning model is constructed based on a back-propagation BP neural network.

In a second aspect, embodiments of the present application also provide a device for determining configuration parameters, including: a first determination module for determining forwarding service simulation information, where the forwarding service simulation information includes input traffic information and data path behavior, where , the input traffic information is used to simulate the input traffic corresponding to multiple forwarding services, the data path behavior is used to simulate the data path that transmits the input traffic; the simulation test module is used to simulate multiple groups based on the data path The first configuration parameters perform simulation tests on the forwarding service simulation information respectively to obtain multiple sets of first simulation test results. The multiple sets of first configuration parameters are different configuration parameters. Among the multiple sets of first simulation test results, Each set of first simulation test results is used to simulate the output traffic after the input traffic corresponding to the multiple forwarding services passes through the data path corresponding to a set of first configuration parameters in the multiple sets of first configuration parameters; second A determining module, configured to determine a target first configuration parameter corresponding to the target first simulation test result from the plurality of sets of first configuration parameters according to the plurality of sets of first simulation test results.

In conjunction with the second aspect, in a first possible implementation of the second aspect, the input traffic information includes any one or more of a service type, a port template, a rate template, a priority template, and a packet length template, The data path behavior includes port information and/or service forwarding paths.

With reference to the first implementation manner of the second aspect, in a second possible implementation manner of the second aspect, the service type includes unicast and/or multicast; the port template includes spontaneous self-collection, half-offset , symmetrical offset, random offset, any one or more of leaf full coverage, leaf front half coverage, leaf back half coverage, leaf interval half coverage and leaf random coverage; the rate template includes burst on, burst Any one or more of unopened, full flow, half flow and custom percentage traffic; the priority template includes one or more traffic transmission priorities, and when the priority template includes multiple traffic transmission In priority, the ratio between the multiple traffic transmission priorities is a preset ratio and/or a custom ratio; the packet length template includes a custom packet length, and the packet length randomly jumps within the preset packet length interval. , the packet length increases and jumps within the preset packet length interval, and any one or more of a variety of packet length mixtures; the port information includes any one of port number, physical address aggregation MAG source, and port rate, or Multiple; the service forwarding path is used to describe the forwarding path and/or forwarding delay of the service.

With reference to any one of the second aspects to the second implementation manner of the second aspect, in a third possible implementation manner of the second aspect, the configuration parameters of the target module on the data path are the same as the The members in the target class corresponding to the target module are assigned values, and different target modules on the data path correspond to different target classes.

With reference to any one of the second aspect to the third implementation manner of the second aspect, in a fourth possible implementation manner of the second aspect, the device further includes: a training module, configured to perform the training according to the Multiple sets of first configuration parameters and the multiple sets of first simulation test results are trained to obtain a machine learning model, and the input of the machine learning model is each set of first configuration parameters in the multiple sets of first configuration parameters. The output of the machine learning model is the first simulation test result corresponding to each set of first configuration parameters.

With reference to the fourth implementation manner of the second aspect, in a fifth possible implementation manner of the second aspect, the device further includes: a third determination module for determining multiple sets of different second configuration parameters; an input module , used to input the multiple sets of different second configuration parameters into the machine learning model respectively to obtain multiple sets of second simulation test results, and the multiple sets of second simulation test results are the output of the machine learning model; Four determination modules, configured to determine a target second simulation test result from the plurality of sets of second simulation test results. The target second simulation test result is consistent with the expected output traffic, and the expected output traffic is the input of the target forwarding service. The output traffic after the traffic passes through the data path; the fifth determination module is used to determine the second configuration parameter corresponding to the target second simulation test result.

In combination with the fourth implementation manner or the fifth implementation manner of the second aspect, in a sixth possible implementation manner of the second aspect, the machine learning model is constructed based on a back-propagation BP neural network.

The device for determining configuration parameters described in the second aspect corresponds to the method for determining configuration parameters described in the first aspect. Therefore, various possible implementations of the second aspect and their beneficial effects can be referred to the corresponding implementations in the first aspect. The relevant descriptions of methods and beneficial effects will not be described in detail here.

In a third aspect, embodiments of the present application further provide a device, including: a processor and a memory; the memory is used to store instructions or computer programs; the processor is used to execute the instructions or computer programs, and execute The method described in any one of the embodiments of the above first aspect.

The device for determining configuration parameters described in the third aspect corresponds to the method for determining configuration parameters described in the first aspect. Therefore, various possible implementations of the third aspect and their beneficial effects can be referred to the corresponding implementations in the first aspect. The relevant descriptions of methods and beneficial effects will not be described in detail here.

In a fourth aspect, embodiments of the present application also provide a computer-readable storage medium, including instructions or computer programs, which, when run on a computer, cause the computer to execute the method described in any one of the embodiments of the first aspect. .

The device for determining configuration parameters described in the fourth aspect corresponds to the method for determining configuration parameters described in the first aspect. Therefore, various possible implementations of the fourth aspect and their beneficial effects can be referred to the corresponding implementations in the first aspect. The relevant descriptions of methods and beneficial effects will not be described in detail here.

In the above implementation manner of the embodiment of the present application, the forwarding service simulation information is first determined. The forwarding service simulation information may include input flow information and data path behavior, wherein the input flow information is used to simulate input corresponding to multiple forwarding services. traffic, and the data path behavior is used to simulate the data path that transmits the input traffic; then, the forwarding service simulation information can be simulated and tested based on multiple sets of configuration parameters of the data path, and multiple sets of simulation test results are obtained. The group configuration parameters are different configuration parameters, and each group of simulation test results among the multiple groups of simulation test results is used to simulate the input traffic corresponding to multiple forwarding services passing through a group of configuration parameters corresponding to the multiple groups of first configuration parameters. The output traffic after the data path is used, so that a target simulation test result can be determined based on multiple sets of simulation test results, and the target configuration parameters corresponding to the target simulation test results can be used as the final required configuration parameters. Since simulation testing is used to determine the configuration parameters corresponding to each module on the data path, the simulation platform can quickly obtain multiple sets of simulation test results based on multiple sets of configuration parameters, so that the required configuration parameters can be quickly determined. On the one hand, it is relatively Compared with repeated trial and error in the on-board test environment, the simulation test method can effectively improve the efficiency of determining configuration parameters and save the overhead required for on-board testing; on the other hand, the simulation platform can traverse multiple sets of configuration parameters and The simulation test results corresponding to each set of configuration parameters are given. It is not necessary for technicians to have high technical analysis capabilities and rich debugging experience. The required configuration parameters can also be determined from multiple sets of configuration parameters, thus reducing the need for technology. Personnel skill requirements. At the same time, the determined configuration parameters can meet the data path performance required when multiple services are overlaid. This can avoid the data path performance of the product only meeting the data path performance in one business scenario but not another business scenario. data path performance.

Description of the drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some implementations recorded in the present application. For example, those of ordinary skill in the art can also obtain other drawings based on these drawings.

Figure 1 is a schematic flow chart of a method for determining configuration parameters in an embodiment of the present application;

Figure 2 is a schematic diagram of defining classes corresponding to modules on the data path in the embodiment of the present application;

Figure 3 is a schematic flow chart of another method for determining configuration parameters in an embodiment of the present application;

Figure 4 is a schematic diagram of the reverse transmission neural network in the embodiment of the present application;

Figure 5 is a schematic structural diagram of a device for determining configuration parameters in an embodiment of the present application;

Figure 6 is a schematic diagram of the hardware structure of a device in an embodiment of the present application.

Detailed ways

During the implementation process of products (such as single boards, chips, etc.), problems with the performance parameters of the product on the data path are often the key pain points in product feature delivery, and the proportion of problems may be more than 60%. In actual applications, there are many combinations of configuration parameters of each module on the data path of the product (such as 300 combinations, etc.). Therefore, it is usually necessary to test the performance parameters on the product data path (that is, each module on the data path) on the test equipment. The configured parameters of the module) are tested on the board to determine which combination of configuration parameters meets the needs of the actual application. When the test results obtained from the on-board test do not meet the technician's expectations, the technician usually adjusts the configured parameters of each module on the data path based on the test results.

However, during the board testing process, it is difficult for technicians to obtain the traffic control data of each module on the data path, and it is difficult to locate the configuration parameters that need to be adjusted on the data path based on the test results. In this case, a high level of knowledge is usually required. Technicians with strong technical analysis capabilities and rich debugging experience have made many adjustments to configuration parameters in the on-board environment so that the test output of the product data path based on the adjusted configuration parameters can meet the expected test requirements. Since the configuration parameters are determined in the on-board test environment, in order to minimize the number of trials and errors, the skill requirements for technical personnel are higher and the difficulty is greater. Moreover, technicians repeatedly try and make mistakes in the on-board test environment, which also reduces the cost. Determine the efficiency of configuration parameters. At the same time, on-board testing involves many areas and material environments, making on-board testing more expensive.

Based on this, embodiments of the present application provide a method for determining configuration parameters, aiming to reduce the skill requirements of technical personnel in the process of determining configuration parameters, improve the efficiency of determining configuration parameters, and reduce overhead. Specifically, the forwarding service simulation information may be determined first. The forwarding service simulation information may include input traffic information and data path behavior. The input flow information is used to simulate the input traffic corresponding to multiple forwarding services, and the data path behavior is used to simulate the input traffic corresponding to multiple forwarding services. to simulate the data path that transmits the input traffic; then, the forwarding service simulation information can be simulated and tested based on multiple sets of configuration parameters of the data path to obtain multiple sets of simulation test results, where the multiple sets of configuration parameters are different configurations parameters, and each set of simulation test results in the multiple sets of simulation test results is used to simulate the output traffic after the input traffic corresponding to the multiple forwarding services passes through the data path corresponding to one set of configuration parameters in the multiple sets of first configuration parameters. , so that a target simulation test result can be determined based on multiple sets of simulation test results. The target simulation test result can be a simulation test result that is consistent with the expected test standard, so that the target configuration parameters corresponding to the target simulation test result can be , as the final required configuration parameters.

Since simulation testing is used to determine the configuration parameters corresponding to each module on the data path, the simulation platform can quickly obtain multiple sets of simulation test results based on multiple sets of configuration parameters, so that the required configuration parameters can be quickly determined. On the one hand, it is relatively Compared with repeated trial and error in the on-board test environment, the simulation test method can effectively improve the efficiency of determining configuration parameters and save the overhead required for on-board testing; on the other hand, the simulation platform can traverse multiple sets of configuration parameters and The simulation test results corresponding to each set of configuration parameters are given. It is not necessary for technicians to have high technical analysis capabilities and rich debugging experience. The required configuration parameters can also be determined from multiple sets of configuration parameters, thus reducing the need for technology. Personnel skill requirements. At the same time, the determined configuration parameters can meet the data path performance required when multiple services are overlaid. This can avoid the data path performance of the product only meeting the data path performance in one business scenario but not another business scenario. data path performance.

In order to make the above objects, features and advantages of the present application more obvious and understandable, various non-limiting implementations in the embodiments of the present application will be illustratively described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

Referring to Figure 1, Figure 1 shows a schematic flow chart of a method for determining configuration parameters in an embodiment of the present application. This method can be applied to a device connected to the simulation platform, or to a device carrying the simulation platform. The method Specific methods may include:

S101: Determine the forwarding service simulation information. The forwarding service simulation information includes input traffic information and data path behavior. The input flow information is used to simulate the input traffic corresponding to the read forwarding service, and the data path behavior is used to simulate the transmission input traffic. data path.

During the simulation test, specific information can be used to simulate multiple forwarding services. Specifically, it is used to simulate the input traffic information and data path behavior of multiple forwarding services on the data path (for the convenience of description, the specific information will be used below. The information is called forwarding service simulation information). Among them, the input traffic information is the corresponding input traffic used to simulate multiple forwarding services for simulation testing, and the data path behavior is the data path used to transmit the input traffic during simulation testing.

During specific implementation, the forwarding service simulation information can be set according to the input flow and output flow of the specific forwarding service on the data path in actual applications. For example, in actual applications, the packet length in the input flow of forwarding service A is usually 64 Bit, then the forwarding service simulation information corresponding to the forwarding service A may include packet length information with a packet length of 64 Bit.

In some exemplary implementations, the data path behavior in forwarding service simulation information may include port information and/or service forwarding paths.

The port information may specifically include any one or more of a port number (such as a port ID), a physical address aggregation (MAC aggregation, MAG) source, and a port rate. The MAG source can be used to describe the LMAG (line MAC aggregation)/FMAG/IMAG source of the port on the hardware; the port rate can be determined based on the network chip, for example, it can be 1Gbps, 10Gbps, 20Gbps, 25Gbps, 40Gbps, 50Gbps, 100Gbps, 200Gbps , 240Gbps, 400Gbps, 500Gbps, etc.

The service forwarding path can be used to describe the path of the forwarded service traffic on the data path (that is, the forwarding path). Optionally, the service forwarding path may also describe the forwarding delay of the forwarded service when forwarding.

In addition, the input traffic information in the forwarding service simulation information may be any one or more of service type, port template, rate template, priority template and packet length template.

Among them, the service type refers to the packet sending method of the forwarding service, which can be divided into two types: unicast and multicast, for example.

The port template can vary according to the service type. Specifically, when the service type is unicast, the port template can include any one or more of spontaneous self-collection, half offset, symmetric offset and random offset. different types of templates. Among them, spontaneous self-receiving means that the destination port (tp) for receiving data packets is the same as the source port (sp) for sending data packets. Half offset refers to the destination port tp of the received data packet=(sp+(n-1)/2)%n, assuming that the number of ports is n and an odd number; or the destination port tp=(sp+n/2)%n , n is the number of ports and is an even number. Symmetric offset refers to the destination port for sending data packets tp=n-1-sp, where n is the number of ports. Random offset means that the destination port tp of the received data packet is any port except the source port sp.

When the service type is multicast, the port template can include any one or more different types of templates, including leaf full coverage, leaf front half coverage, leaf back half coverage, leaf interval half coverage, and leaf random coverage. Full leaf coverage means that the number of multicast leaf members is equal to the number of ports, that is, each multicast leaf member corresponds to a port and serves as the destination port for receiving data packets.

The first half of the leaf coverage means that the multicast leaf members correspond to the first half of the ports or approximately the first half of the ports. For example, assume that the number of ports is n, numbered from 0 to n respectively. If n is an odd number, the destination port tp for receiving the data packet (that is, the destination port corresponding to the multicast leaf member) is {0,1,2 ,…,(n-1)/2}, if n is an even number, the destination port tp is {0,1,2,…,n/2-1}.

Correspondingly, the second half of the leaf coverage means that the multicast leaf members correspond to the second half of the ports or approximately the last half of the ports in a one-to-one manner. For example, it is still assumed that the number of ports is n, numbered from 0 to n respectively, and n is an odd number, then the destination port tp for receiving the data packet is {(n-1)/2,...,n-1}, if n is an even number , then the destination port tp is {n/2,…,n-1}.

Leaf interval half coverage means that multicast leaf members correspond to half of the ports or approximately half of the ports, and there is at least one port between the ports corresponding to any two multicast leaf members. For example, assuming the number of ports is n, numbered from 0 to n respectively, then if n is an odd number, the destination port tp for receiving the data packet can be {0,2,4,…,n-1}, if n is an even number , then the destination port tp can be {0,2,4,…,n-2}.

Leaf random coverage refers to randomly selecting the destination port corresponding to the multicast leaf member from the total ports. For example, assuming that the number of ports is n, numbered from 0 to n respectively, and the number of multicast leaf members is m (m≤n), then tp is m ports randomly selected from n ports.

The rate template in the input traffic information can include any one or more of burst on, burst off, full flow, half flow, and custom percentage flow.

Among them, burst opening refers to allowing the port traffic to change suddenly, for example, allowing the port traffic to suddenly increase to a larger value or suddenly decrease to a smaller value, etc. Correspondingly, bursting is not enabled, which means that the traffic on the port is not allowed to change suddenly. In actual applications, the change value of the port traffic can be compared with the preset threshold to determine whether a burst occurs.

Full traffic means that the transmission rate of data packets (traffic) of the forwarding service at the port reaches the maximum transmission rate of the port. Half traffic means that the transmission rate of forwarding service data packets at the port reaches one-half of the maximum transmission rate corresponding to the port. The custom percentage means that the transmission rate of the data packets of the forwarding service at the port can be set by the user (technical personnel). For example, it can be set to 40%, which indicates the transmission rate of the data packets of the forwarding service at the port. Reach 40% of the maximum transmission rate corresponding to the port.

The priority template in the input traffic information can include one or more traffic transmission priorities, which represent the priority of forwarding service traffic when it is transmitted at the port. For example, assuming that the priority corresponding to the traffic of forwarding service A is Pri0, and the priority corresponding to the traffic of forwarding service B is Pri7, then when the traffic of forwarding service A and the traffic of forwarding service B are transmitted by the same port at the same time, priority can be The traffic corresponding to forwarding service B is transmitted, and then the traffic corresponding to forwarding service A is transmitted (assuming that the larger the priority value, the higher the priority level, and the more priority the traffic is transmitted).

Assume that the priority of the traffic transmitted at the port has pri0, pri1, ..., pri7, a total of 8 types (of course, in other implementations, there can be more or less than 8 types, etc.), then the priority template can Contains one traffic transmission priority, such as pri0 priority, or pri7 priority, etc.; it can also contain multiple traffic transmission priorities, such as including both pri0 and pri7, and the two traffic transmission priorities are The traffic ratio can be 1:1 or other preset ratios, or it can include 8 traffic transmission priorities from pri0 to pri7, and the traffic ratio between each traffic transmission priority can be a preset ratio, such as 1:1 :1:1:1:1:1:1 etc. In a further possible implementation, when the priority template includes multiple traffic transmission priorities, the ratio between the various traffic transmission priorities can also be a ratio customized by the user (technical personnel). For example, the traffic of forwarding service A can be Including traffic with two traffic transmission priorities, Pri0 and Pri7, the user can define the traffic ratio of these two traffic transmission priorities to be 3:7, etc.

The packet length template in the input traffic information refers to the length template of the data packets in the input traffic of the forwarding service, which can include customized packet lengths, packet lengths that randomly jump within the preset packet length interval, and packet lengths within the preset packet length interval. Incremental jumps within long intervals and any one or more of a variety of packet length mixtures.

Among them, the custom packet length, that is, the length of the data packet in the input traffic of the forwarding service can be set by the user. For example, the user can set the length of each data packet in the input traffic of the forwarding service to 64Bit, or to 128Bit etc.

The packet length jumps randomly within the preset packet length interval, which means that the length of the data packet in the input traffic of the forwarding service can be any value (positive integer) in the preset packet length interval. For example, assuming that the default packet length interval is [64Bit, 256Bit], or [64Bit, 9600Bit], the packet length can be 64Bit, 128Bit, 192Bit, 256Bit and other length values. Moreover, the packet lengths of different data packets can be different.

The packet length jumps incrementally within the preset packet length interval, which means that the length of each data packet in the input traffic of the forwarding service can gradually increase, and the packet length of each data packet is within the preset packet length interval. For example, assuming that the default packet length range is [64Bit, 256Bit], or [64Bit, 9600Bit], the packet length can start from 64Bit and increase sequentially. Assuming that the increment step size is 1Bit, the packet length of the first data packet can be 64Bit, the packet length of the second data packet can be 65Bit, the packet length of the third data packet can be 66Bit, etc. Of course, in actual applications, when the packet length of a certain data packet is 256 Bit, the packet length of the next data packet can be incremented from 64 Bit again.

Mixing of multiple packet lengths means that data packets of different packet lengths in the input traffic of the forwarding service can be mixed according to a certain proportion. For example, assuming that forwarding service A traffic contains data packets with two packet lengths, 64Bit and 1518Bit, the ratio of these two packet lengths can be 7:3, that is, 70% of the data packets have a packet length of 64Bit. The packet length of 30% of the data packets is 1518Bit. In other examples, the forwarding service A traffic contains data packets with four packet lengths: 64Bit, 78Bit, 576Bit, and 1518Bit, and the ratios are 55:5:17:23; or the forwarding service A traffic contains 64Bit, 130Bit , 260Bit, 577Bit, 1518Bit and 9000Bit six packet lengths, the ratios are 449:160:200:80:110:1, etc.

Optionally, for each forwarding service in actual applications, in this embodiment, triplet information may be used to simulate the forwarding service. As input to the simulation test, the triplet information can specifically be {port information, input traffic information, service forwarding path}. In this way, when a port includes input traffic of multiple forwarding services, the triplet information corresponding to each forwarding service (or the above-mentioned business simulation information) can be used to characterize the service flows of multiple forwarding services under the port, so that it can Realize the overlay of multiple forwarding business scenarios, and then based on the configuration parameters determined in the overlay business scenario, can meet the data path performance requirements in multiple different forwarding business scenarios.

S102: Perform simulation tests on the forwarding service simulation information based on multiple sets of first configuration parameters of the data path, and obtain multiple sets of first simulation test results, where the multiple sets of first configuration parameters are different configuration parameters, and multiple sets of first configuration parameters are Each set of first simulation test results in the first simulation test results is used to simulate the output traffic after input traffic corresponding to multiple forwarding services passes through a data path corresponding to a set of first configuration parameters in multiple sets of first configuration parameters. .

For example, after determining the forwarding service simulation information corresponding to each forwarding service, the forwarding service simulation information can be input into a simulation platform for simulation testing; the simulation platform can be based on multiple sets of different configuration parameters of the data path. Perform a simulation test on the forwarding service simulation information to obtain the simulation test results corresponding to each set of configuration parameters. Each group of configuration parameters is a combination of configuration parameters corresponding to each module on the data path. Configuration parameters in different groups correspond to different configuration parameters of at least one module on the data path. For convenience of description, this configuration parameter is called the first configuration parameter below. Each set of simulation test results obtained is used to simulate the output traffic after input traffic corresponding to multiple forwarding services passes through the data path corresponding to the corresponding first configuration parameter. For the convenience of description, this simulation test result is referred to as the first simulation test result below.

It can be understood that through simulation testing, the first simulation test results corresponding to each set of first configuration parameters can be obtained, thereby eliminating the need to test the traffic output corresponding to the set of first configuration parameters in the board environment. In this way, the efficiency of obtaining test results is usually relatively high, and the hardware overhead required for on-board testing can also be saved.

In an exemplary specific implementation, during the simulation test process, classes can be used to abstractly describe corresponding modules on the data path. Specifically, the data path or the single board/chip carrying the data path can be used as a major category, and each module on the data path can be used as an internal class. Correspondingly, the sub-modules under each module can be As a subclass of this internal class, the public attributes of each module can be used as members of the parent class, and the private attributes of each module can be used as members of the subclass. In this way, the parameter configuration process for each module on the data path is the process of assigning values to members in the corresponding classes for the simulation platform. Take the target module on the data path as an example. The target module can be any module on the data path. Then configuring the parameters of the target module is to configure the members of the class corresponding to the target module in the simulation platform. Make an assignment. Among them, different target modules on the data path can correspond to different target classes, and the configuration parameters of different target modules correspond to member assignments in different target classes.

As an example, according to the module on the data path, the class corresponding to the module can be defined as shown in Figure 2. Take the four-level waterline configuration parameters of Total level, RES level, Channel_Group level, and Channel level of the RB module as an example. Specifically, the corresponding members in the RB class, RES class, CHAN_GROUP class, and CHAN class can be assigned values. Similarly, the process of configuring parameters for the PA module and PORT module is similar to the process of configuring the parameters of the EB module, and will not be described in detail here.

In this way, the simulation platform can assign values to the corresponding class members according to each group of first configuration parameters, thereby simulating the parameter configuration process of each module on the data path, and simulate the input forwarding business information based on the assigned class members. The simulation test is performed to obtain corresponding first simulation test results, so that multiple sets of first simulation test results can be based on different sets of first configuration parameters, and each set of first configuration parameters corresponds to at least one set of first simulation test results.

It is worth noting that the simulation platform in this embodiment may be an electronic system level (ESL) simulation platform or other simulation platforms, which is not limited in this embodiment.

S103: According to the obtained multiple sets of first simulation test results, the target first configuration parameter corresponding to the target first simulation test result can be determined from the multiple sets of first configuration parameters.

In practical applications, after obtaining multiple sets of first simulation test results, the first simulation test result that is consistent with the expected traffic output result can be determined (for convenience of description, hereafter referred to as the target first simulation test result), And a set of first configuration parameters used to obtain the target first simulation test result will be used as the final target first configuration parameters that are expected to be determined.

It can be seen that the above-mentioned process of determining the target first configuration parameter through simulation testing is relatively more efficient in determining the target first configuration parameter than the process in which technicians conduct testing in a board environment to determine the target first configuration parameter. , Moreover, it can effectively save the hardware environment overhead required for on-board testing. At the same time, the simulation platform can automatically conduct simulation tests based on multiple sets of different first configuration parameters and provide corresponding first simulation test results. It does not require technicians to have high technical analysis capabilities and rich debugging experience, which reduces the time required for technicians. skill requirements. In addition, since the platform can perform simulation tests on the forwarding service simulation information of multiple forwarding services, this can enable the finally determined first configuration parameter to meet the performance requirements for the data path in a variety of different forwarding service scenarios.

In practical applications, after multiple sets of first simulation test results are determined based on multiple sets of different first configuration parameters based on the simulation platform, machine learning can also be performed based on multiple sets of first configuration parameters and multiple sets of first simulation test results. . In this way, there is no need to use a simulation platform to conduct simulation tests. The model obtained by machine learning can be used to quickly determine the simulation test results corresponding to each set of configuration parameters, thereby further reducing the work input of data path designers, saving labor costs, and improving Determine the efficiency of configuration parameters.

Referring to Figure 3, Figure 3 shows a schematic flowchart of another method for determining configuration parameters in an embodiment of the present application. After executing steps S101 to S103 as shown in Figure 1, the method can be further executed:

Step S104: Train to obtain a machine learning model based on multiple sets of first configuration parameters and multiple sets of first simulation test results. The input of the machine learning model is each set of first configuration parameters in the multiple sets of first configuration parameters. The machine learning model The output of is the first simulation test result corresponding to each set of first configuration parameters.

In this embodiment, the first configuration parameters used by the simulation platform for simulation testing can be recorded, as well as the first simulation test results obtained based on the set of first configuration parameters, which can then be divided into training and testing machine learning models. sample set and test set.

In the process of training the machine learning model, each set of first configuration parameters in the sample set can be used as the input of the machine learning model, and the first simulation test result corresponding to the set of first configuration parameters can be used as the output of the machine learning model, The machine learning model is trained to obtain the machine learning model by using multiple sets of first configuration parameters and multiple sets of first simulation test results in the sample set.

Then, each set of first configuration parameters included in the test set and its corresponding first simulation test result can be used to test the trained machine learning model. Specifically, each set of first configuration parameters in the test set is used as the input of the machine learning model, the machine learning model outputs a result based on the input configuration parameters, and the result is corresponding to the first configuration parameter in the test set. Compare the first simulation test results. If the results output by the machine learning model and the corresponding first simulation test results meet the preset test requirements (for example, the similarity reaches the preset standard), it can be determined that the machine learning model has passed Test, and if the output results of the machine learning model and the corresponding simulation test results do not meet the preset test requirements, the machine learning model can continue to be trained.

As an example, the machine learning model can be built based on a back propagation (BP) neural network. For example, the BP neural network can be shown in Figure 4, which includes an input layer, a hidden layer and an output layer. Assume that the input traffic of the forwarding service includes at most 7 traffic transmission priorities from Pri0 to Pri6. Each group of first configuration parameters includes n parameters, which are Param ₀ , Param ₁ , Param ₂ ,..., Param _n-1. , then when training the machine learning model, each set of first configuration parameters Param ₀ , Param ₁ , Param ₂ ,..., Param _n-1 can be sent to the input layer as the sample input of the model, and priority will be given to Pri0 to Pri7 in the bandwidth. The discard percentages LossPercent ₀ , LossPercent ₁ , LossPercent ₂ ,..., LossPercent ₇ (that is, the simulation test results) corresponding to the level of traffic in sequence are used as the sample output of the BP neural network to train the machine learning model (or BP neural network).

In this way, after the machine learning model is trained, there is no need to rely on a simulation platform for simulation testing. For each forwarding business scenario or a scenario where multiple forwarding services are superimposed, the machine learning model can be used to quickly output according to each set of configuration parameters. The simulation test results corresponding to the set of configuration parameters are obtained, thereby further improving the efficiency of determining the configuration parameters. In specific implementation, the configuration parameters may be determined by continuing to execute steps S105 to S107.

S105: Determine multiple sets of different second configuration parameters.

After training the machine learning model, if you need to determine the configuration parameters that meet the required data path performance in a certain business scenario, you can use the machine learning model to obtain the simulation test results corresponding to each set of configuration parameters. For the convenience of description, the configuration parameters input into the machine learning model are referred to as second configuration parameters below, and the configuration parameters that need to be finally determined are one set of configuration parameters in the second configuration parameters. The multiple sets of different second configuration parameters determined may be given by technical personnel, or all possible configuration parameter combinations of each module on the data path may be traversed, and all or part of them may be used as the configuration parameters in this embodiment. Second configuration parameter.

In a further possible implementation, the expected traffic output after the input traffic of the target forwarding service (any one or more services in actual applications) passes through the data path can also be determined. Normally, the input traffic of the target forwarding service passes through After the configuration parameters of each module on the data path are adjusted, the output traffic on the data path is consistent with the expected traffic output, which means that the configuration parameters of each module on the data path can enable the performance of the data path to meet the target forwarding business scenario. requirements. Therefore, the expected traffic output after the input traffic of the target forwarding service passes through the data path can also be determined.

S106: Input the multiple sets of different second configuration parameters into the machine learning model respectively to obtain multiple sets of second simulation test results, where the multiple sets of second simulation test results are outputs of the machine learning model.

When obtaining the simulation test results, the second configuration parameters can be input into the machine learning model, and the machine learning model outputs the second simulation test results corresponding to each set of second configuration parameters, so that based on the multiple sets of second configuration parameters, we can obtain Multiple sets of second simulation test results.

S107: Determine the target second simulation test result from multiple sets of second simulation test results, where the target second simulation test result is consistent with the expected output traffic, and the expected output traffic is the input traffic of the target forwarding service after passing through the data path. output flow.

In specific implementation, after using the machine learning model to obtain multiple sets of simulation test results, each second simulation test result can be compared with the expected traffic output after the input traffic of the target forwarding service passes through the data path. If multiple sets of second simulation test results are If the traffic output represented by the target second simulation test result is consistent with the expected traffic output in the test results, then the candidate configuration parameters corresponding to the target second simulation test result are the configuration parameters that need to be determined.

S108: Determine the second configuration parameter corresponding to the target second simulation test result.

After the target second simulation test result is determined based on step S107, when the machine learning model outputs the target second simulation test result, the second configuration parameter as the input of the machine learning model is determined as the configuration parameter that needs to be determined, that is, That is, the configuration parameters meet the data path performance requirements required by the target forwarding business scenario.

In the above process, since the machine learning model can be used to obtain the simulation test results corresponding to each set of configuration parameters, it is no longer necessary to obtain the required simulation test results corresponding to each set of configuration parameters through simulation testing, so that it is possible to Improving the efficiency of obtaining simulation test results can further improve the efficiency of determining configuration parameters.

In addition, the embodiment of the present application also provides a device for determining configuration parameters. Referring to Figure 5, Figure 5 shows a schematic structural diagram of a device for determining configuration parameters in an embodiment of the present application. The device 500 includes:

The first determination module 501 is used to determine the forwarding service simulation information. The forwarding service simulation information includes input flow information and data path behavior, wherein the input flow information is used to simulate the input flow corresponding to multiple forwarding services, so The data path behavior described above is used to simulate the data path transmitting the input traffic;

The simulation test module 502 is configured to conduct simulation tests on the forwarding service simulation information based on multiple sets of first configuration parameters of the data path, and obtain multiple sets of first simulation test results. The multiple sets of first configuration parameters are different Configuration parameters, each set of first simulation test results in the plurality of first simulation test results is used to simulate the input traffic corresponding to the plurality of forwarding services passing through a group of first configuration parameters in the plurality of sets. The output traffic after the data path corresponding to the configuration parameter;

The second determination module 503 is configured to determine the target first configuration parameter corresponding to the target first simulation test result from the multiple sets of first configuration parameters according to the multiple sets of first simulation test results.

In a possible implementation, the input traffic information includes any one or more of service type, port template, rate template, priority template and packet length template, and the data path behavior includes port information and/ or service forwarding path.

In a possible implementation, the service type includes unicast and/or multicast;

The port template includes any one or more of spontaneous self-closing, half offset, symmetrical offset, random offset, full leaf coverage, leaf front half coverage, leaf back half coverage, leaf interval half coverage, and leaf random coverage. ;

The rate template includes any one or more of burst on, burst off, full flow, half flow, and custom percentage flow;

The priority template includes one or more traffic transmission priorities, and when the priority template includes multiple traffic transmission priorities, the ratio between the multiple traffic transmission priorities is a preset ratio and/ Or custom proportion;

The packet length template includes any one or more of a custom packet length, a random jump of the packet length within a preset packet length interval, an incremental jump of the packet length within the preset packet length interval, and a mixture of multiple packet lengths. ;

The port information includes any one or more of port number, physical address aggregation MAG source, and port rate;

The service forwarding path is used to describe the forwarding path and/or forwarding delay of the service.

In a possible implementation, the configuration parameters of the target module on the data path are assigned to members in the target class corresponding to the target module, and different target modules on the data path correspond to different target classes.

In a possible implementation, the device 500 further includes:

A training module, configured to train and obtain a machine learning model based on the plurality of sets of first configuration parameters and the plurality of sets of first simulation test results, where the input of the machine learning model is each of the plurality of sets of first configuration parameters. A set of first configuration parameters, and the output of the machine learning model is a first simulation test result corresponding to each set of first configuration parameters.

In a possible implementation, the device 500 further includes:

The third determination module is used to determine multiple sets of different second configuration parameters;

An input module, configured to input the multiple sets of different second configuration parameters into the machine learning model to obtain multiple sets of second simulation test results, where the multiple sets of second simulation test results are outputs of the machine learning model. ;

The fourth determination module is configured to determine a target second simulation test result from the plurality of sets of second simulation test results, the target second simulation test result is consistent with the expected output traffic, and the expected output traffic is the target forwarding service. The output traffic after the input traffic passes through the data path;

The fifth determination module is used to determine the second configuration parameter corresponding to the target second simulation test result.

In a possible implementation, the machine learning model is constructed based on a back-propagation BP neural network.

It should be noted that the information interaction, execution process, etc. between the modules of the above-mentioned device are based on the same concept as the method embodiments in the embodiments of this application, and the technical effects they bring are the same as those of the method embodiments in the embodiments of this application. , For the convenience and simplicity of description, for the specific working processes of the above-described devices and modules, reference can be made to the corresponding processes in the foregoing method embodiments, which will not be described again here.

In addition, the embodiment of the present application also provides a device that can perform the method of determining configuration parameters described in the above method embodiment. Referring to Figure 6, Figure 6 shows an exemplary hardware structure diagram of the device. The device can be the simulation platform in any of the above embodiments or a device carrying the simulation platform. This device is a hardware structure device that can be used to implement the functional modules in the device 500 shown in Figure 5. For example, those skilled in the art can imagine that the first determination module 501, the simulation test module 502 and the second determination module 503 in the device 500 shown in FIG. 5 can be implemented by at least one processor 601 calling the code in the memory 602. Optionally, the device can also be used to implement the functions of the simulation platform or the device carrying the simulation platform in any of the above embodiments.

The device may include at least one processor 601 and at least one memory 602. The processor 601 may be connected to the memory 602, for example, through a bus 603 as shown in FIG. 6 . Of course, in actual applications, the connection between the processor 601 and the memory 602 may include various interfaces, transmission lines or buses, etc., which is not limited in this embodiment.

The memory 602 may be used to store instructions or computer programs;

The processor 601 can be used to execute the instructions or computer programs, and perform the following steps according to the computer programs or instructions:

Determine the forwarding service simulation information, the forwarding service simulation information includes input traffic information and data path behavior, wherein the input flow information is used to simulate the input traffic corresponding to multiple forwarding services, and the data path behavior is used to simulate transmission The data path of the input traffic;

Simulation tests are performed on the forwarding service simulation information based on multiple sets of first configuration parameters of the data path to obtain multiple sets of first simulation test results. The multiple sets of first configuration parameters are different configuration parameters. The multiple sets of first configuration parameters are different configuration parameters. Each group of first simulation test results in the group of first simulation test results is used to simulate the input traffic corresponding to the plurality of forwarding services passing through the data corresponding to a group of first configuration parameters in the plurality of groups of first configuration parameters. Output flow after the passage;

According to the plurality of sets of first simulation test results, a target first configuration parameter corresponding to the target first simulation test result is determined from the plurality of sets of first configuration parameters.

In some possible implementations, the input traffic information includes any one or more of service type, port template, rate template, priority template, and packet length template, and the data path behavior includes port information and/or Service forwarding path.

In some possible implementations, the service type includes unicast and/or multicast;

The port template includes any one or more of spontaneous self-closing, half-offset, symmetrical offset, random offset, full leaf coverage, leaf front half coverage, leaf back half coverage, leaf interval half coverage, and leaf random coverage. ;

The rate template includes any one or more of burst on, burst off, full flow, half flow, and custom percentage flow;

The priority template includes one or more traffic transmission priorities, and when the priority template includes multiple traffic transmission priorities, the ratio between the multiple traffic transmission priorities is a preset ratio and/ Or custom proportion;

The packet length template includes any one or more of a custom packet length, a random jump of the packet length within a preset packet length interval, an incremental jump of the packet length within the preset packet length interval, and a mixture of multiple packet lengths. ;

The port information includes any one or more of port number, physical address aggregation MAG source, and port rate;

The service forwarding path is used to describe the forwarding path and/or forwarding delay of the service.

In some possible implementations, the configuration parameters of the target module on the data path assign values to members in the target class corresponding to the target module, and different target modules on the data path correspond to different target classes.

In some possible implementations, the processor 601 can also perform the following steps according to the computer program or instructions:

According to the multiple sets of first configuration parameters and the multiple sets of first simulation test results, a machine learning model is trained, and the input of the machine learning model is each set of first configuration parameters in the multiple sets of first configuration parameters. , the output of the machine learning model is the first simulation test result corresponding to each set of first configuration parameters.

In some possible implementations, the processor 601 can also perform the following steps according to the computer program or instructions:

Determine multiple sets of different second configuration parameters;

Input the multiple sets of different second configuration parameters into the machine learning model respectively to obtain multiple sets of second simulation test results, and the multiple sets of second simulation test results are the output of the machine learning model;

A target second simulation test result is determined from the plurality of sets of second simulation test results. The target second simulation test result is consistent with the expected output traffic. The expected output traffic is the input traffic of the target forwarding service passing through the data path. The final output flow;

Determine the second configuration parameter corresponding to the target second simulation test result.

In some possible implementations, the machine learning model is constructed based on a back-propagation BP neural network.

It is worth noting that the operation content executed by the processor 601 in the above device according to the computer program or instructions stored in the memory 602 is based on the same concept as the method embodiments in the embodiments of the present application, and the technical effects it brings are the same as those in the embodiments of the present application. The method embodiments are the same. For the convenience and simplicity of description, for the specific working process of the processing execution steps described above, reference can be made to the corresponding processes in the foregoing method embodiments, which will not be described again here.

It should be noted that the processor in this application may include but is not limited to at least one of the following: central processing unit (CPU), microprocessor, digital signal processing (DSP), microcontroller (microcontroller unit, MCU), or artificial intelligence processor and other types of computing devices that run software. Each computing device may include one or more cores for executing software instructions to perform calculations or processing. The processor can be a separate semiconductor chip, or it can be integrated with other circuits into a semiconductor chip. For example, it can form an on-chip system (such as a codec circuit, a hardware acceleration circuit, or various buses and interface circuits) with other circuits (such as codec circuits, hardware acceleration circuits, or various bus and interface circuits). system-on-a-chip), or can be integrated into the ASIC as a built-in processor of an application-specific integrated circuit (ASIC). The ASIC with integrated processor can be packaged separately or can be combined with Other circuits are packaged together. In addition to a core for executing software instructions to perform calculations or processing, the processor may further include necessary hardware accelerators, such as field programmable gate array (FPGA), programmable logic device , PLD), or a logic circuit that implements dedicated logic operations.

The memory in the embodiment of the present application may include at least one of the following types: read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (random access memory, RAM) or other types of dynamic storage devices that can store information and instructions, or electrically erasable programmable-only memory (EEPROM). In some scenarios, the memory can also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage medium, magnetic disk storage medium or other magnetic storage device, or can be used for portability or storage Any other medium that has the desired program code in the form of instructions or data structures and can be accessed by a computer, without limitation.

The memory can exist independently and be connected to the processor. Among them, the memory can store program codes for executing the technical solutions of the embodiments of the present application, and the execution is controlled by the processor. Various types of computer program codes that are executed can also be regarded as drivers of the processor. For example, the processor is used to execute the computer program code stored in the memory, thereby implementing the technical solutions in the embodiments of the present application.

In addition, embodiments of the present application also provide a computer-readable storage medium, which includes instructions or computer programs that, when run on a computer, cause the computer to perform the method of determining configuration parameters described in the above method embodiment.

In this application, "of", corresponding "(corresponding, relevant)" and "corresponding (English: corresponding)" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the intended The meaning of the expression is consistent.

It should be noted that in the embodiments of this application, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "such as" in the embodiments of the present application is not to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the words "exemplary" or "such as" is intended to present the concept in a concrete manner.

In this application, "at least one" means one or more. "Plural" means two or more. "And/or" describes the association of associated objects, indicating that there can be three relationships, for example, A and/or B, which can mean: A exists alone, A and B exist simultaneously, and B exists alone, where A, B can be singular or plural. The character "/" generally indicates that the related objects are in an "or" relationship. "At least one of the following" or similar expressions thereof refers to any combination of these items, including any combination of a single item (items) or a plurality of items (items). For example, at least one of a, b, or c can mean: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, c can be single or multiple . In addition, in order to facilitate a clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, words such as “first” and “second” are used to distinguish identical or similar items with basically the same functions and effects. Those skilled in the art can understand that words such as "first" and "second" do not limit the number and execution order, and words such as "first" and "second" do not limit the number and execution order.

The system architecture and business scenarios described in the embodiments of this application are for the purpose of explaining the technical solutions of the embodiments of this application more clearly, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Persons of ordinary skill in the art will know that with the network With the evolution of architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

In the several embodiments provided in this application, it should be understood that the disclosed devices, equipment and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or software functional units.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program code. .

As mentioned above, the above embodiments are only used to illustrate the technical solution of the present application, but not to limit it. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still make the foregoing technical solutions. The technical solutions described in each embodiment may be modified, or some of the technical features may be equivalently replaced; however, these modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the scope of the technical solutions in each embodiment of the present application.

Claims (14)

1. A method for determining configuration parameters, characterized in that the method includes: Determine the forwarding service simulation information, the forwarding service simulation information includes input traffic information and data path behavior, wherein the input flow information is used to simulate the input traffic corresponding to multiple forwarding services, and the data path behavior is used to simulate transmission The data path of the input traffic; Simulation tests are performed on the forwarding service simulation information based on multiple sets of first configuration parameters of the data path to obtain multiple sets of first simulation test results. The multiple sets of first configuration parameters are different configuration parameters. The multiple sets of first configuration parameters are different configuration parameters. Each group of first simulation test results in the group of first simulation test results is used to simulate the input traffic corresponding to the plurality of forwarding services passing through the data corresponding to a group of first configuration parameters in the plurality of groups of first configuration parameters. Output traffic after the path, wherein the configuration parameters of the target module on the data path are assigned to members in the target class corresponding to the target module, the target module is a module on the data path, and the data Different modules on the pathway correspond to different classes; According to the plurality of sets of first simulation test results, a target first configuration parameter corresponding to the target first simulation test result is determined from the plurality of sets of first configuration parameters. 2. The method according to claim 1, characterized in that the input traffic information includes any one or more of service type, port template, rate template, priority template and packet length template, and the data path Behavior includes port information and/or service forwarding paths. 3. The method according to claim 2, characterized in that, The service type includes unicast and/or multicast; The port template includes any one or more of spontaneous self-closing, half-offset, symmetrical offset, random offset, full leaf coverage, leaf front half coverage, leaf back half coverage, leaf interval half coverage, and leaf random coverage. ; The rate template includes any one or more of burst on, burst off, full flow, half flow, and custom percentage flow; The priority template includes one or more traffic transmission priorities, and when the priority template includes multiple traffic transmission priorities, the ratio between the multiple traffic transmission priorities is a preset ratio and/ Or custom proportion; The packet length template includes any one or more of a custom packet length, a random jump of the packet length within a preset packet length interval, an incremental jump of the packet length within the preset packet length interval, and a mixture of multiple packet lengths. ; The port information includes any one or more of port number, physical address aggregation MAG source, and port rate; The service forwarding path is used to describe the forwarding path and/or forwarding delay of the service. 4. The method according to any one of claims 1 to 3, characterized in that the method further includes: According to the multiple sets of first configuration parameters and the multiple sets of first simulation test results, a machine learning model is trained, and the input of the machine learning model is each set of first configuration parameters in the multiple sets of first configuration parameters. , the output of the machine learning model is the first simulation test result corresponding to each set of first configuration parameters. 5. The method according to claim 4, characterized in that, the method further comprises: Determine multiple sets of different second configuration parameters; Input the multiple sets of different second configuration parameters into the machine learning model respectively to obtain multiple sets of second simulation test results, and the multiple sets of second simulation test results are the output of the machine learning model; A target second simulation test result is determined from the plurality of sets of second simulation test results. The target second simulation test result is consistent with the expected output traffic. The expected output traffic is the input traffic of the target forwarding service passing through the data path. The final output flow; Determine the second configuration parameter corresponding to the target second simulation test result. 6. The method according to claim 4, characterized in that the machine learning model is constructed based on a back propagation BP neural network. 7. A device for determining configuration parameters, characterized in that the device includes: The first determination module is used to determine the forwarding service simulation information. The forwarding service simulation information includes input flow information and data path behavior, wherein the input flow information is used to simulate the input flow corresponding to multiple forwarding services. The data path behavior is used to simulate the data path carrying the input traffic; A simulation test module, configured to perform simulation tests on the forwarding service simulation information based on multiple sets of first configuration parameters of the data path to obtain multiple sets of first simulation test results, where the multiple sets of first configuration parameters are different Configuration parameters, each set of first simulation test results in the plurality of sets of first simulation test results is used to simulate that the input traffic corresponding to the plurality of forwarding services passes through a set of first among the plurality of sets of first configuration parameters. The output traffic after configuring the data path corresponding to the parameter, wherein the configuration parameters of the target module on the data path are assigned values to the members in the target class corresponding to the target module, and the target module is on the data path. A module, different modules on the data path correspond to different classes; The second determination module is configured to determine the target first configuration parameter corresponding to the target first simulation test result from the plurality of sets of first configuration parameters according to the plurality of sets of first simulation test results. 8. The device according to claim 7, wherein the input traffic information includes any one or more of service type, port template, rate template, priority template and packet length template, and the data path Behavior includes port information and/or service forwarding paths. 9. The device according to claim 8, characterized in that, The service type includes unicast and/or multicast; The port template includes any one or more of spontaneous self-closing, half-offset, symmetrical offset, random offset, full leaf coverage, leaf front half coverage, leaf back half coverage, leaf interval half coverage, and leaf random coverage. ; The rate template includes any one or more of burst on, burst off, full flow, half flow, and custom percentage flow; The priority template includes one or more traffic transmission priorities, and when the priority template includes multiple traffic transmission priorities, the ratio between the multiple traffic transmission priorities is a preset ratio and/ Or custom proportion; The packet length template includes any one or more of a custom packet length, a random jump of the packet length within a preset packet length interval, an incremental jump of the packet length within the preset packet length interval, and a mixture of multiple packet lengths. ; The port information includes any one or more of port number, physical address aggregation MAG source, and port rate; The service forwarding path is used to describe the forwarding path and/or forwarding delay of the service. 10. The device according to any one of claims 7 to 9, characterized in that the device further includes: A training module, configured to train and obtain a machine learning model based on the plurality of sets of first configuration parameters and the plurality of sets of first simulation test results, where the input of the machine learning model is each of the plurality of sets of first configuration parameters. A set of first configuration parameters, and the output of the machine learning model is a first simulation test result corresponding to each set of first configuration parameters. 11. The device according to claim 10, characterized in that the device further comprises: The third determination module is used to determine multiple sets of different second configuration parameters; An input module, configured to input the multiple sets of different second configuration parameters into the machine learning model to obtain multiple sets of second simulation test results, where the multiple sets of second simulation test results are outputs of the machine learning model. ; The fourth determination module is configured to determine a target second simulation test result from the plurality of sets of second simulation test results, the target second simulation test result is consistent with the expected output traffic, and the expected output traffic is the target forwarding service. The output traffic after the input traffic passes through the data path; The fifth determination module is used to determine the second configuration parameter corresponding to the target second simulation test result. 12. The device according to claim 10, wherein the machine learning model is constructed based on a back propagation BP neural network. 13. A device for determining configuration parameters, characterized by comprising: a processor and a memory; The memory is used to store instructions or computer programs; The processor is configured to execute the instructions or computer program to perform the method described in any one of claims 1-6. 14. A computer-readable storage medium, characterized in that it includes instructions or computer programs that, when run on a computer, cause the computer to perform the method described in any one of claims 1-6 above.