Configuration optimization method and related equipment of service grid system

Abstract

The present invention provides a configuration optimization method and related equipment for a service grid system, belonging to the field of communications technology. The method comprises: obtaining current configuration parameters and target performance of the service grid system; starting the service grid system using the current configuration parameters, and testing the started service grid system using an automatic testing program to obtain first test data; obtaining a knowledge graph corresponding to the first test parameters; and optimizing the configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance. In the present invention, the test data is obtained by testing the service grid system configured with the current configuration parameters, and the service grid system is automatically optimized based on the test data, the knowledge graph, and the target performance, without the need for manual optimization. This avoids the problem of excessive optimization time caused by manual optimization of the service grid system, shortens the optimization time of the service grid system's configuration parameters, and ensures the robustness of the service grid system after it goes online.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a configuration optimization method and related equipment for a service grid system. Background Art

[0002] There are many components in a distributed system, and each component is configured with multiple modules and multiple units, resulting in a large number of modules and units deployed in the distributed system.

[0003] The normal operation of modules and units in a distributed system is monitored by the service mesh system. This monitoring system is configured to monitor the modules and units in the distributed system. However, the monitoring system sets configuration parameters based on rules, limiting the service mesh system to exceptions within the scope of historical experience. This can result in performance that falls short of expectations after the service mesh system is launched.

[0004] To ensure the expected performance of a service mesh system, its configuration parameters need to be optimized. Currently, manual optimization of service mesh system configuration parameters is time-consuming, and service mesh systems have a launch deadline, making it difficult to ensure robustness after launch. Summary of the Invention

[0005] The present invention provides a configuration optimization method and related equipment for a service grid system, which are used to solve the problem that the configuration parameter optimization time of the service grid system is long.

[0006] In one aspect, the present invention provides a configuration optimization method for a service grid system, comprising:

[0007] Get the current configuration parameters and target performance of the service grid system;

[0008] Starting the service grid system using the current configuration parameters, and testing the started service grid system through an automatic testing program to obtain first test data;

[0009] Obtaining a knowledge graph corresponding to the first test parameter, the knowledge graph including a mapping relationship between test data and performance;

[0010] Optimize configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance.

[0011] In one embodiment, the step of optimizing the configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance includes:

[0012] Determining the current performance of the service grid system based on the first test data and the knowledge graph;

[0013] When the current performance does not match the target performance, optimizing configuration parameters of the service grid system according to the current performance and the target performance.

[0014] In one embodiment, the step of optimizing the configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance includes:

[0015] Determining the current performance of the service grid system based on the first test data and the knowledge graph;

[0016] Determining a quality score and a poor quality score of the current configuration parameters according to the target performance and the current performance;

[0017] Optimizing configuration parameters of the service grid system according to the high-quality score and the low-quality score.

[0018] In one embodiment, the step of optimizing the configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance includes:

[0019] Determining the current performance of the service grid system based on the first test data and the knowledge graph;

[0020] Determining an item to be optimized of the service grid system according to the target performance and the current performance;

[0021] Optimize the configuration parameters of the item to be optimized.

[0022] In one embodiment, the step of optimizing the configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance includes:

[0023] Inputting the first test data, the knowledge graph, and the target performance into a target generative adversarial network model to obtain a high-quality score and a low-quality score for the current configuration parameters;

[0024] Optimizing configuration parameters of the service grid system according to the high-quality score and the low-quality score.

[0025] In one embodiment, before the step of inputting the first test data and the target performance into a target generative adversarial network model to obtain a high-quality score and a low-quality score for the current configuration parameters, the step further includes:

[0026] Obtaining a normal sample, where the normal sample is determined based on the knowledge graph;

[0027] Generate a random vector, and control a generator in the generative adversarial network model to be trained to generate an abnormal sample based on the random vector;

[0028] The discriminator in the to-be-trained generative adversarial network model is controlled to discriminate between the abnormal samples and the normal samples, so as to train the to-be-trained generative adversarial network model to obtain the target generative adversarial network model.

[0029] In another aspect, the present invention provides a service grid system, comprising:

[0030] The acquisition module is used to obtain the current configuration parameters and target performance of the service grid system;

[0031] a startup module, configured to start the service grid system using the current configuration parameters, and test the started service grid system through an automatic testing program to obtain first test data;

[0032] The acquisition module is further configured to acquire a knowledge graph corresponding to the first test parameter, wherein the knowledge graph includes a mapping relationship between test data and performance;

[0033] An optimization module is used to optimize the configuration parameters of the service grid system based on the first test data, the knowledge graph and the target performance, so that the performance of the service grid system with the optimized configuration parameters reaches the target performance.

[0034] In another aspect, the present invention provides a service grid system, comprising: a memory and a processor;

[0035] The memory stores computer-executable instructions;

[0036] The processor executes the computer-executable instructions stored in the memory, so that the processor performs the configuration optimization method of the service grid system as described above.

[0037] On the other hand, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the configuration optimization method of the service grid system as described above.

[0038] On the other hand, the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the configuration optimization method of the service grid system as described above.

[0039] The present invention provides a configuration optimization method and related equipment for a service grid system. The method obtains the current configuration parameters and target performance of the service grid system, starts the service grid system using the current configuration parameters, tests the started service grid system using an automatic testing program to obtain test data, and then obtains a knowledge graph corresponding to the test data. Based on the test data, the knowledge graph, and the target performance, the configuration parameters of the service grid system are optimized, so that the performance of the service grid system with the optimized configuration parameters reaches the target performance. In the present invention, the service grid system is automatically optimized using the test data, the knowledge graph corresponding to the test data, and the target performance, without the need for manual optimization. This avoids the problem of excessive optimization time caused by manual optimization of the service grid system, significantly shortens the service grid system configuration parameter optimization time, and ensures the robustness of the service grid system after it goes online. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0041] Figure 1 A system architecture diagram of a configuration optimization method for a service grid system implemented by the present invention;

[0042] Figure 2 This is a flow chart of a first embodiment of a configuration optimization method for a service grid system according to the present invention;

[0043] Figure 3 This is a detailed flowchart of step S40 in the second embodiment of the configuration optimization method for the service grid system of the present invention;

[0044] Figure 4 This is a detailed flowchart of step S40 in the third embodiment of the configuration optimization method for the service grid system of the present invention;

[0045] Figure 5 This is a detailed flowchart of step S40 in the fourth embodiment of the configuration optimization method for the service grid system of the present invention;

[0046] Figure 6 This is a detailed flowchart of step S40 in the fifth embodiment of the configuration optimization method for the service grid system of the present invention;

[0047] Figure 7 This is a module diagram of the service grid system of the present invention;

[0048] Figure 8 Schematic diagram of the hardware structure of the service grid system of the present invention.

[0049] The above drawings illustrate specific embodiments of the present disclosure, which will be described in more detail below. These drawings and textual descriptions are not intended to limit the scope of the present disclosure in any way, but rather to illustrate the concepts of the present disclosure to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0050] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

[0051] The present invention provides a configuration optimization method for a service grid system, which can be Figure 1 The system architecture diagram shown is implemented. Figure 1 As shown, service mesh system 100 includes multiple meshes 110, each of which is configured to monitor a component 210 in distributed system 200. That is, each mesh 110 interacts with the corresponding component 210. Service mesh system 100 must achieve target performance to monitor each component 210. Therefore, before service mesh system 100 goes online to monitor distributed system 200, its configuration parameters must be optimized to ensure that service mesh system 100 approaches or achieves the target performance.

[0052] The following describes in detail the technical solutions of the present invention and how the technical solutions of this application solve the above-mentioned technical problems using specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. The following embodiments of the present invention are described in conjunction with the accompanying drawings.

[0053] Reference Figure 2 , Figure 2 This is a first embodiment of a configuration optimization method for a service grid system according to the present invention. The configuration optimization method for a service grid system includes the following steps:

[0054] Step S10: Obtain the current configuration parameters and target performance of the service grid system.

[0055] In this embodiment, when configuring the configuration parameters of the service grid system, the service grid system's goals and resources are set, that is, the target performance of the service grid is set. For example, the target performance of the service grid system is: the current resources of the service grid system can support 1000 concurrencies with a response time of less than 100ms. The target performance can be determined by the distributed system that the service grid system needs to monitor. For example, if the number of components, modules, and units that need to be monitored in a distributed system is 1000, then the maximum concurrency of the service grid system should be set to 1000, that is, the service grid system can simultaneously receive a maximum of 1000 pieces of data that need to be processed from the distributed system. For another example, if the distributed system has a latency requirement of less than 100ms, then the response time of the service grid system should be less than 100ms. The target performance can be written into the service grid system, so that the service grid system can achieve the target performance.

[0056] Each grid in a service grid system is configured with a corresponding service. When optimizing configuration parameters, the service grid system needs to obtain all parameters that affect its configuration, including both discrete and continuous parameters. For example, the service grid system obtains grid configuration parameters and service configuration parameters. The configuration parameters obtained by the service grid system are defined as the current configuration parameters.

[0057] Step S20: Start the service grid system using the current configuration parameters, and test the started service grid system through an automatic testing program to obtain first test data.

[0058] When optimizing a service grid system, it is necessary to optimize based on the current performance of the service grid system. The service grid system will start up using the current configuration parameters, and the current performance of the service grid system started up using the current configuration parameters can be measured. Specifically, the service grid system is provided with an automatic testing program. After the service grid system is started up, the automatic testing program tests the functions of the service grid system to obtain test data. The obtained test data is defined as first test data. The first test data can represent the performance of the service grid system configured with the current configuration parameters. The functions tested by the automatic testing program can be simple access to an API (Application Programming Interface), complex access to system functions of the entire service grid system, or performance testing.

[0059] It should be noted that before testing the service grid system, it is necessary to verify that the basic functional cabinets of the service grid system are functioning properly. In other words, functional testing of the service grid system must first confirm that the functions are functioning properly, so that first test data representing the current performance of the service grid system can be accurately obtained. First test data may include heartbeat data, network data, log data, system data, indicator data, etc.

[0060] Step S30: Obtain a knowledge graph corresponding to the first test parameter, where the knowledge graph includes a mapping relationship between test data and performance.

[0061] The service grid system stores a knowledge graph. This graph includes mappings between test data and performance. This knowledge graph is derived from the experience of experts and R&D personnel.

[0062] After receiving the first test data, the service grid system obtains the knowledge graph corresponding to the first test data. Specifically, the service grid system stores multiple knowledge graphs, and the types of test data in the knowledge graphs are associated with the knowledge graphs. After receiving the first test data, the service grid system determines the type of the first test data and, based on the type, obtains the knowledge graph corresponding to the first test data.

[0063] Step S40: Optimize the configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance.

[0064] After obtaining the knowledge graph, the service grid system can optimize the configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance. Specifically, the service grid system can score the current configuration parameters of the service grid system using the knowledge graph, the first test data, and the target performance. If the score is low, the service grid system parameters need to be reconfigured for optimization. For example, the service grid system determines the current performance of the service grid system based on the knowledge graph and the first test data, measures the current performance and the target performance, and then determines the score based on the ratio of the measured current performance to the target performance.

[0065] In the technical solution provided in this embodiment, the current configuration parameters and target performance of the service grid system are obtained, the service grid system is started using the current configuration parameters, and the started service grid system is tested using an automatic testing program to obtain test data. The knowledge graph corresponding to the test data is then obtained, and the configuration parameters of the service grid system are optimized based on the test data, the knowledge graph, and the target performance, so that the performance of the service grid system with the optimized configuration parameters reaches the target performance. In the present invention, test data is obtained by testing the service grid system configured with the current configuration parameters, and the service grid system is automatically optimized based on the test data, the knowledge graph corresponding to the test data, and the target performance, without the need for manual optimization. This avoids the problem of excessive optimization time caused by manual optimization of the service grid system, greatly shortens the service grid system configuration parameter optimization time, and ensures the robustness of the service grid system after it goes online.

[0066] Reference Figure 3 , Figure 3This is a second embodiment of the configuration optimization method for a service grid system of the present invention. Based on the first embodiment, step S40 includes:

[0067] Step S401: Determine the current performance of the service grid system based on the first test data and the knowledge graph.

[0068] Step S402: When the current performance does not match the target performance, the configuration parameters of the service grid system are optimized according to the current performance and the target performance.

[0069] In this embodiment, the knowledge graph includes a mapping relationship between test data and performance, so the service grid system can determine the current performance of the service grid system based on the knowledge graph and the first test data.

[0070] The service grid performance then determines whether the current performance matches the target performance, that is, whether the current performance is close to or reaches the target performance. If the gap between the current performance and the target performance is large, it can be determined that the current performance does not match the target performance. For example, the current performance is a 4-core CPU with 16G memory, which can support a maximum of 1000 concurrency; but the target performance needs to support 2000 concurrency, then the gap between the current performance and the target performance is large. The difference between the target performance and the value converted by the current performance is greater than the preset threshold, and the gap between the current performance and the target performance is large. At this time, it is necessary to optimize the configuration parameters of the service grid system based on the current performance and the target performance. For example, based on the current performance and the target performance, it is determined that the concurrency supported by the service grid system is low, then it is necessary to optimize the configuration parameters in the service grid system that can increase the concurrency.

[0071] In the technical solution provided in this embodiment, the service grid system determines the current performance of the service grid system based on the first test data and the knowledge graph. If the current performance does not match the target performance, the configuration parameters of the service grid system are optimized based on the current performance and the target performance.

[0072] Reference Figure 4 , Figure 4 This is a third embodiment of the configuration optimization method for a service grid system of the present invention. Based on the first embodiment, step S40 includes:

[0073] Step S403: Determine the current performance of the service grid system based on the first test data and the knowledge graph.

[0074] Step S404: Determine the quality score and the poor quality score of the current configuration parameters according to the target performance and the current performance.

[0075] Step S405: Optimize configuration parameters of the service grid system based on the high-quality scores and the low-quality scores.

[0076] In this embodiment, the knowledge graph includes a mapping relationship between test data and performance, so the service grid system can determine the current performance of the service grid system based on the knowledge graph and the first test data.

[0077] After determining the current performance, the service mesh system determines the quality scores and quality scores for the current configuration parameters based on the target performance and the current performance. The quality score refers to the ratio of configuration sub-parameters in the service mesh system that do not require optimization to the current configuration parameters, while the quality score refers to the ratio of configuration sub-parameters in the service mesh system that require optimization to the current configuration parameters. For example, if the quality score is 30 and the quality score is 70, it means that 30% of the configuration sub-parameters in the service mesh system do not require optimization, while 70% do.

[0078] The current configuration parameter has multiple configuration subparameters, each corresponding to a current subperformance. The current subperformance is compared with the target performance of the same type. If the two do not match, the unmatched configuration subparameter is incremented by 1; if they do match, the matching configuration subparameter is incremented by 1. The service grid system then calculates the ratio of the number of matching configuration subdata to the total number of all configuration subdata, which is the quality score. The service grid system calculates the ratio of the number of mismatching configuration subdata to the total number of all configuration subdata, which is the poor quality score.

[0079] The service grid system can optimize its configuration parameters based on the quality and quality scores. Specifically, the configuration sub-parameters corresponding to the quality scores are optimized, and the configuration sub-parameters corresponding to the quality scores are adjusted accordingly based on the optimized configuration sub-parameters to ensure improved overall performance of the service grid system. For example, if a first configuration sub-parameter requires optimization, while a second configuration sub-parameter that does not require optimization is linked to the first configuration sub-parameter, then the second configuration sub-parameter also needs to be adjusted based on the optimized first configuration sub-parameter.

[0080] In the technical solution provided in this embodiment, the service grid system determines the current performance of the service grid system based on the first test data and the knowledge graph, and thus determines the quality score and poor quality score of the current configuration parameters based on the current performance and the target performance. Finally, the configuration parameters of the service grid system are accurately optimized through the quality score and poor quality score.

[0081] Reference Figure 5 , Figure 5 This is a fourth embodiment of the configuration optimization method for a service grid system of the present invention. Based on the first embodiment, step S40 includes:

[0082] Step S406: Determine the current performance of the service grid system based on the first test data and the knowledge graph.

[0083] Step S407: Determine the items to be optimized in the service grid system based on the target performance and the current performance.

[0084] Step S408: Optimize the configuration parameters of the item to be optimized.

[0085] In this embodiment, the knowledge graph includes a mapping relationship between test data and performance, so the service grid system can determine the current performance of the service grid system based on the knowledge graph and the first test data.

[0086] The service grid system can use target performance and target performance to identify items that need to be optimized, also known as optimization targets. For example, if the current performance response time is 200ms and the target performance is 100ms, all functions or applications that affect response time are identified as optimization targets.

[0087] After determining the items to be optimized, the service grid system can optimize the configuration parameters of the items to be optimized, thereby improving the performance of the service grid system and enabling the service grid system to achieve the target performance.

[0088] In the technical solution provided in this embodiment, the items to be optimized of the service grid system can be determined based on the current performance and target performance of the service grid system, thereby accurately optimizing the configuration parameters of the items to be optimized.

[0089] Reference Figure 6 , Figure 6 This is the fifth embodiment of the service grid system of the present invention, based on the first embodiment, step S40 includes:

[0090] Step S409: Input the first test data, the knowledge graph, and the target performance into the target generative adversarial network model to obtain the quality score and the poor quality score of the current configuration parameters.

[0091] Step S410: Optimize configuration parameters of the service grid system based on the high-quality scores and the low-quality scores.

[0092] Generative adversarial networks (GANs) are highly efficient unsupervised (or semi-supervised) learning algorithms, primarily consisting of a generator and a verifier. The generator generates a large number of query questions, which are then effectively verified by training the verifier to produce a GAN model. After sufficient iterations, the GAN model develops strong anomaly detection capabilities and can identify configuration parameters that require optimization for the service mesh system based on detected anomalies.

[0093] In this embodiment, a generative adversarial network model is stored in the service grid system, which is defined as a target generative adversarial network model. The service grid system inputs the first test data, the knowledge graph, and the target performance into the target generative adversarial network model. The target generative adversarial network model makes a judgment based on the input parameters and outputs a high-quality score and a low-quality score. The high-quality score and the low-quality score are the high-quality score and the low-quality score of the current configuration parameters. The service grid system can optimize the configuration of the service grid system using the high-quality score and the low-quality score. The explanation of the high-quality score and the low-quality score, and the optimization of the configuration parameters of the service grid system using the high-quality score and the low-quality score are specifically described above and will not be repeated here.

[0094] The service grid system needs to train a generative adversarial network model to obtain a target generative adversarial network model. Specifically, normal samples from the service grid system are obtained. Normal samples are real data and can be determined through a knowledge graph. The generative adversarial network model is equipped with a noise device that generates random vectors. The generator of the generative adversarial network model generates abnormal samples based on the random vectors. The discriminator of the generative adversarial network model distinguishes between abnormal samples and normal samples to train the generative adversarial network model to obtain the target generative adversarial network.

[0095] That is, before the step of inputting the first test data and the target performance into the target generative adversarial network model to obtain the quality score and the poor quality score of the current configuration parameters, the step further includes:

[0096] Obtain normal samples, which are determined based on the knowledge graph;

[0097] Generate a random vector and control the generator in the generative adversarial network model to be trained to generate abnormal samples based on the random vector;

[0098] The discriminator in the generative adversarial network model to be trained is controlled to distinguish abnormal samples from normal samples, so as to train the generative adversarial network model to be trained to obtain the target generative adversarial network model.

[0099] During the training process, the generator and discriminator are trained alternately. First, the generator is fixed. It simulates G(z) based on the latent random vector z as a negative sample. Positive samples x are sampled from real data (which can be test data in a knowledge graph). These positive and negative samples are then fed into the discriminator for binary classification. Finally, the discriminator parameters are updated using the binary cross-entropy loss. The discriminator is then fixed and optimized. To minimize the chance of misleading the discriminator, i.e., to ensure that the discriminator classifies the generated "fake" samples as positive, the generator is typically optimized to maximize the probability of the generated samples being classified as positive.

[0100] In the technical solution provided in this embodiment, the service grid system accurately determines its own anomalies by generating an adversarial network model, and then accurately optimizes its own configuration parameters based on the anomalies.

[0101] The present invention also provides a service grid system, referring to Figure 7 , the service grid system 700 includes:

[0102] Acquisition module 710, used to obtain current configuration parameters and target performance of the service grid system;

[0103] A startup module 720 is configured to start the service grid system using current configuration parameters and test the started service grid system using an automatic testing program to obtain first test data;

[0104] An acquisition module 710 is configured to acquire a knowledge graph corresponding to the first test parameter, where the knowledge graph includes a mapping relationship between test data and performance.

[0105] The optimization module 730 is used to optimize the configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance.

[0106] In one embodiment, the service grid system 700 includes:

[0107] A determination module, configured to determine the current performance of the service grid system based on the first test data and the knowledge graph;

[0108] The optimization module 730 is used to optimize the configuration parameters of the service grid system according to the current performance and the target performance when the current performance does not match the target performance.

[0109] In one embodiment, the service grid system 700 includes:

[0110] A determination module, configured to determine the current performance of the service grid system based on the first test data and the knowledge graph;

[0111] A determination module is used to determine the quality score and poor quality score of the current configuration parameters based on the target performance and the current performance;

[0112] The optimization module 730 is used to optimize the configuration parameters of the service grid system based on the high-quality scores and the low-quality scores.

[0113] In one embodiment, the service grid system 700 includes:

[0114] A determination module, configured to determine the current performance of the service grid system based on the first test data and the knowledge graph;

[0115] The determination module is used to determine the items to be optimized in the service grid system based on the target performance and current performance;

[0116] The optimization module 730 is used to optimize the configuration parameters of the item to be optimized.

[0117] In one embodiment, the service grid system 700 includes:

[0118] An input module is used to input the first test data, the knowledge graph, and the target performance into the target generative adversarial network model to obtain a high-quality score and a low-quality score of the current configuration parameters;

[0119] The optimization module 730 is used to optimize the configuration parameters of the service grid system based on the high-quality scores and the low-quality scores.

[0120] In one embodiment, the service grid system 700 includes:

[0121] An acquisition module 710 is used to acquire normal samples, where the normal samples are determined based on the knowledge graph;

[0122] A generation module is used to generate random vectors and control the generator in the generative adversarial network model to be trained to generate abnormal samples based on the random vectors;

[0123] The control module is used to control the discriminator in the to-be-trained generative adversarial network model to discriminate between abnormal samples and normal samples, so as to train the to-be-trained generative adversarial network model to obtain a target generative adversarial network model.

[0124] Figure 8 The figure is a hardware structure diagram of a service grid system according to an exemplary embodiment.

[0125] The service grid system 800 may include: a processor 801, such as a CPU, a memory 802, and a transceiver 803. Those skilled in the art will appreciate that Figure 8 The structure shown in the figure does not limit the service grid system and may include more or fewer components than shown, or a combination of certain components or a different arrangement of components. Memory 802 may be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0126] The processor 801 may call a computer program stored in the memory 802 to complete all or part of the steps of the above-mentioned method for optimizing the configuration of the service grid system.

[0127] The transceiver 803 is used to receive information sent by an external device and send information to the external device.

[0128] A non-transitory computer-readable storage medium, when instructions in the storage medium are executed by a processor of a service grid system, enables the service grid system to perform the above-mentioned service grid system configuration optimization method.

[0129] A computer program product includes a computer program. When the computer program is executed by a processor of a service grid system, the service grid system is enabled to perform the configuration optimization method of the service grid system.

[0130] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0131] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A configuration optimization method for a service grid system, characterized in that: include: Obtaining current configuration parameters and target performance of the service grid system; the current configuration parameters include multiple configuration sub-parameters, each configuration sub-parameter corresponding to a current sub-performance; Starting the service grid system using the current configuration parameters, and testing the started service grid system through an automatic testing program to obtain first test data; Obtaining a knowledge graph corresponding to the first test data, the knowledge graph including a mapping relationship between the test data and performance; Optimizing configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance, including: determining a current performance of the service grid system based on the first test data and the knowledge graph; and optimizing the configuration parameters of the service grid system based on the current performance of the service grid system and the target performance; Optimizing configuration parameters of the service grid system according to the current performance of the service grid system and the target performance includes: Comparing each of the current sub-performances with the same type of sub-performances in the target performance to determine the number of matching configuration sub-parameters and the number of unmatched configuration sub-parameters; Determining a high-quality score of the current configuration parameter based on a ratio of the number of the matching configuration sub-parameters to the total number of all configuration sub-parameters, and determining a low-quality score of the current configuration parameter based on a ratio of the number of the unmatched configuration sub-parameters to the total number of all configuration sub-parameters; The configuration parameters of the service grid system are optimized based on the high-quality scores and the low-quality scores. That is, the configuration sub-parameters corresponding to the low-quality scores are optimized, and the configuration sub-parameters corresponding to the high-quality scores are also adjusted accordingly based on the optimized configuration sub-parameters.

2. The configuration optimization method of the service grid system according to claim 1, characterized in that: Optimizing configuration parameters of the service grid system according to the current performance of the service grid system and the target performance includes: When the current performance does not match the target performance, optimizing configuration parameters of the service grid system according to the current performance and the target performance.

3. The configuration optimization method of the service grid system according to claim 1, characterized in that: Optimizing configuration parameters of the service grid system according to the current performance of the service grid system and the target performance includes: Determining an item to be optimized of the service grid system according to the target performance and the current performance; Optimize the configuration parameters of the item to be optimized.

4. The configuration optimization method of the service grid system according to claim 1, characterized in that: Optimizing the configuration parameters of the service grid system according to the first test data, the knowledge graph, and the target performance includes: Inputting the first test data, the knowledge graph, and the target performance into a target generative adversarial network model to obtain a high-quality score and a low-quality score for the current configuration parameters; Optimizing configuration parameters of the service grid system according to the high-quality score and the low-quality score.

5. The configuration optimization method of the service grid system according to claim 4, characterized in that: Before the step of inputting the first test data, the knowledge graph, and the target performance into a target generative adversarial network model to obtain a high-quality score and a low-quality score of the current configuration parameters, the method further includes: Obtaining a normal sample, where the normal sample is determined based on the knowledge graph; Generate a random vector, and control a generator in the to-be-trained generative adversarial network model to generate an abnormal sample based on the random vector; The discriminator in the to-be-trained generative adversarial network model is controlled to discriminate between the abnormal samples and the normal samples, so as to train the to-be-trained generative adversarial network model to obtain the target generative adversarial network model.

6. A service grid system, characterized in that: include: An acquisition module, configured to acquire current configuration parameters and target performance of the service grid system; the current configuration parameters include multiple configuration sub-parameters, each configuration sub-parameter corresponding to a current sub-performance; a startup module, configured to start the service grid system using the current configuration parameters, and test the started service grid system through an automatic testing program to obtain first test data; The acquisition module is further configured to acquire a knowledge graph corresponding to the first test data, wherein the knowledge graph includes a mapping relationship between the test data and the performance; An optimization module is configured to optimize configuration parameters of the service grid system based on the first test data, the knowledge graph, and the target performance, including: determining a current performance of the service grid system based on the first test data and the knowledge graph; optimizing the configuration parameters of the service grid system based on the current performance of the service grid system and the target performance; wherein optimizing the configuration parameters of the service grid system based on the current performance of the service grid system and the target performance includes: comparing each of the current sub-performances with a sub-performance of the same type in the target performance to determine the number of matching configuration sub-parameters and the number of unmatched configuration sub-parameters; determining a high-quality score of the current configuration parameter based on a ratio of the number of matching configuration sub-parameters to the total number of all configuration sub-parameters, and determining a low-quality score of the current configuration parameter based on a ratio of the number of unmatched configuration sub-parameters to the total number of all configuration sub-parameters; and optimizing the configuration parameters of the service grid system based on the high-quality score and the low-quality score, i.e., optimizing the configuration sub-parameters corresponding to the low-quality score and adjusting the configuration sub-parameters corresponding to the high-quality score based on the optimized configuration sub-parameters.

7. A service grid system, characterized in that: include: Memory and processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor executes the configuration optimization method for the service grid system according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the configuration optimization method for the service grid system according to any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the configuration optimization method of the service grid system according to any one of claims 1 to 5 is implemented.

Images