CN-122019609-A - Performance analysis optimization method based on AI and related equipment thereof

CN122019609ACN 122019609 ACN122019609 ACN 122019609ACN-122019609-A

Abstract

The application belongs to the technical field of artificial intelligence, and relates to an AI-based performance analysis optimization method and related equipment thereof, wherein multi-source isomerization performance analysis data are acquired from a target system; analyzing to identify system performance defects, inputting the system performance defects into a generated large language model, generating a system performance optimization scheme according to system performance defect repair or optimization data in the generated large language model, analyzing the system performance optimization scheme, and performing performance optimization on a target system. Through fully utilizing the AI processing assembly in the data acquisition, data analysis and subsequent system performance optimization scheme generation processes, a great amount of manpower consumption is saved, manual analysis steps are reduced, and the accuracy of analysis results can be ensured as much as possible. The performance analysis optimization method is used for system performance data analysis, particularly in a financial business scene with larger business data volume and relatively more impurities, and can remarkably improve analysis efficiency and discover business performance abnormality in time.

Inventors

TIAN JUAN
LI JIANQIANG

Assignees

平安科技（深圳）有限公司

Dates

Publication Date: 20260512
Application Date: 20260113

Claims (10)

1. The AI-based performance analysis optimization method is characterized by comprising the following steps: collecting performance analysis data of multi-source isomerization from a target system; Identifying a system performance defect by analyzing the performance analysis data; inputting the system performance defects into a generated large language model, and generating a system performance optimization scheme according to system performance defect repair or optimization data screened by the generated large language model; Analyzing the system performance optimization scheme, and performing performance optimization on the target system.
2. The AI-based performance analysis optimization method of claim 1, wherein the step of collecting performance analysis data of the multi-source isomerization from the target system comprises: collecting the performance analysis data from a target system by using a buried point mode, wherein the multi-source heterogeneous performance analysis data comprises application log data, system monitoring index data and link logic tracking data; After performing the step of collecting performance analysis data for multi-source isomerization from the target system, the method further comprises: and preprocessing the performance analysis data, wherein the preprocessing comprises data format unified conversion and data cleaning.
3. The AI-based performance analysis optimization method according to claim 1 or 2, wherein the step of identifying a system performance defect by analyzing the performance analysis data specifically includes: performing unified time sequence feature vector conversion on the performance analysis data; determining a target execution unit or a target execution function when the performance index is abnormal by carrying out association analysis on the unified time sequence feature vector; And identifying the system performance defect according to the target execution unit or the target execution function.
4. The AI-based performance analysis optimization method of claim 3, wherein the step of performing unified timing feature vector conversion on the performance analysis data specifically comprises: establishing an association mapping relation between the performance analysis data, the target source code file and the target code row according to the acquisition path of the performance analysis data; And performing logic alignment and normalization conversion processing on the performance analysis data according to the association mapping relation and the execution time stamp corresponding to the target source code to construct a unified time sequence feature vector containing the system performance state, the business application behavior and the code context.
5. The AI-based performance analysis optimization method of claim 3, wherein the step of determining the target execution unit or the target execution function when the performance index is abnormal by performing the association analysis on the unified timing feature vector specifically includes: Inputting the unified time sequence feature vector into a pre-trained association analysis model, and identifying strong correlation features for representing performance index abnormality, wherein the pre-trained association analysis model comprises an association analysis model based on a Pelson correlation coefficient, and the strong correlation features refer to features with correlation exceeding a preset correlation threshold; Screening out specific log errors, system resource bottleneck nodes and high-delay nodes which cause abnormal performance indexes based on the strong correlation characteristics, wherein the high-delay nodes refer to nodes with delay time exceeding a preset time threshold; And positioning the specific log errors, the system resource bottleneck nodes and the high-delay nodes to a target execution unit or a target execution function.
6. The AI-based performance analysis optimization method of claim 3, wherein the identifying the system performance defect based on the target execution unit or the target execution function specifically includes: and adopting an analytical large language model to carry out static analysis on the source codes related to the target execution unit or the target execution function, and identifying the system performance defect.
7. The AI-based performance analysis optimization method according to claim 6, wherein the step of inputting the system performance defect into a generative large language model, and generating a system performance optimization scheme according to system performance defect repair or optimization data screened by the generative large language model specifically comprises the steps of: Taking a performance defect description field corresponding to the system performance defect as an optimization prompt word; Inputting the optimized Prompt word into the generated large language model, and acquiring a Prompt template output by the generated large language model; And screening out the step-by-step optimization processing step with the highest confidence from the system performance defect repair or optimization data, and gradually adding the step-by-step optimization processing step into the Prompt template to generate the system performance optimization scheme.
8. An AI-based performance analysis optimizing apparatus, comprising: The data acquisition module is used for acquiring performance analysis data of multi-source isomerization from the target system; the data analysis module is used for identifying system performance defects by analyzing the performance analysis data; The optimization scheme generating module is used for inputting the system performance defects into a generated large language model, and generating a system performance optimization scheme according to system performance defect repairing or optimizing data screened by the generated large language model; And the performance optimization execution module is used for analyzing the system performance optimization scheme and performing performance optimization on the target system.
9. A computer device comprising a memory having stored therein computer readable instructions which when executed by the processor implement the steps of the AI-based performance analysis optimization method of any of claims 1-7.
10. A computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the AI-based performance analysis optimization method of any of claims 1 to 7.

Description

Performance analysis optimization method based on AI and related equipment thereof Technical Field The application relates to the technical field of artificial intelligence, which is applied to a system performance analysis optimization scene, and relates to an AI-based performance analysis optimization method and related equipment thereof. Background Currently, the Internet industry is currently using various types of performance testing tools and application performance management tools, such as sampling performance metrics, such as response or throughput, at regularly spaced time intervals by an Information Technology (IT) management tool. Some IT management tools detect performance anomalies by setting thresholds for various performance metrics, e.g., anomalies are detected when a performance metric exceeds or falls below a specified threshold. In this way, manual analysis, which is highly dependent on a senior architect or performance engineer, can be used in small scale business systems. However, if the method is adopted for the financial business system with larger number of clients and more frequent interaction, a great amount of time is required to understand the root cause of the problem, and the problem is found to be solved by self-designed based on the root cause, so that the period from the problem finding to the problem solving is too long, the abnormal performance of the system cannot be timely processed, the optimization of the system performance cannot be timely performed under serious conditions, and serious financial loss is caused, so that how to quickly and timely perform the analysis and optimization of the system performance becomes a problem to be solved urgently. Disclosure of Invention The embodiment of the application aims to provide an AI-based performance analysis optimization method and related equipment thereof, so as to solve the technical problem of how to quickly and timely perform system performance analysis optimization in the prior art. In a first aspect, an embodiment of the present application provides an AI-based performance analysis optimization method, which adopts the following technical scheme: The AI-based performance analysis optimization method comprises the following steps: collecting performance analysis data of multi-source isomerization from a target system; Identifying a system performance defect by analyzing the performance analysis data; Inputting the system performance defects into a generated large language model, and generating a system performance optimization scheme according to system performance defect repair or optimization data screened by the generated large language model, wherein the generated large language model comprises a code generated model based on a transducer architecture; Analyzing the system performance optimization scheme, and performing performance optimization on the target system. In a second aspect, an embodiment of the present application further provides an AI-based performance analysis optimization apparatus, which adopts the following technical scheme: An AI-based performance analysis optimizing apparatus comprising: The data acquisition module is used for acquiring performance analysis data of multi-source isomerization from the target system; the data analysis module is used for identifying system performance defects by analyzing the performance analysis data; The optimization scheme generation module is used for inputting the system performance defects into a generation type large language model, and generating a system performance optimization scheme according to system performance defect repair or optimization data screened by the generation type large language model, wherein the generation type large language model comprises a code generation type model based on a transducer architecture; And the performance optimization execution module is used for analyzing the system performance optimization scheme and performing performance optimization on the target system. In a third aspect, an embodiment of the present application further provides a computer device, which adopts the following technical scheme: A computer device comprising a memory having stored therein computer readable instructions which when executed by the processor implement the steps of the AI-based performance analysis optimization method described above. In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, which adopts the following technical solutions: a computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the AI-based performance analysis optimization method as described above. Compared with the prior art, the embodiment of the application has the following main beneficial effects: the AI-based performance analysis optimization method comprises the steps of collecting multi-source isomerization perfor