Performance analysis optimization method based on AI and related equipment thereof
By collecting multi-source heterogeneous data from financial business systems and utilizing AI-generated large language models, performance defects are automatically identified and optimized, solving the problem of excessively long performance analysis and optimization cycles in financial business systems and achieving efficient and accurate system performance optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- PING AN TECH (SHENZHEN) CO LTD
- Filing Date
- 2026-01-13
- Publication Date
- 2026-05-12
AI Technical Summary
In existing financial business systems with a large number of customers and frequent interactions, the performance analysis and optimization cycle is too long, which makes it impossible to handle performance anomalies in a timely manner and may even cause significant losses.
An AI-based performance analysis and optimization method is adopted. By collecting multi-source heterogeneous performance analysis data from the target system, a generative large language model is used to identify system performance defects and generate optimization solutions. This includes data collection, analysis, and optimization execution modules, and automated analysis and optimization are performed using a code generation model based on the Transformer architecture.
It significantly improves the efficiency of system performance analysis, reduces manual analysis steps, ensures the accuracy of analysis results, and enables timely detection and resolution of business performance anomalies, especially in financial business scenarios with large and complex business data.
Smart Images
Figure CN122019609A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology and is applied to system performance analysis and optimization scenarios, involving AI-based performance analysis and optimization methods and related equipment. Background Technology
[0002] Currently, the internet industry widely uses various performance testing and application performance management tools. For example, IT management tools sample performance metrics such as response time or throughput at regular intervals. Some IT management tools detect performance anomalies by setting thresholds for various performance metrics; for instance, anomalies are detected when performance metrics exceed or fall below specified thresholds. This approach heavily relies on manual analysis by senior architects or performance engineers and is suitable for small-scale business systems.
[0003] However, for financial business systems with a large number of customers and frequent interactions, adopting the above method would require a significant amount of time to understand the root cause of the problem and design solutions based on that cause. This would result in an excessively long cycle from "discovering the problem" to "solving the problem," which would not only lead to the inability to handle system performance anomalies in a timely manner, but in severe cases, the inability to optimize system performance in a timely manner could also cause significant financial losses. Therefore, how to quickly and timely perform system performance analysis and optimization has become an urgent problem to be solved. Summary of the Invention
[0004] The purpose of this application is to propose an AI-based performance analysis and optimization method and related equipment to solve the technical problem of how to quickly and timely perform system performance analysis and optimization in the prior art.
[0005] Firstly, embodiments of this application provide an AI-based performance analysis and optimization method, which employs the following technical solutions: AI-based performance analysis and optimization methods include: Collect multi-source heterogeneous performance analysis data from the target system; By analyzing the performance analysis data, system performance defects are identified; The system performance defects are input into a generative large language model. Based on the system performance defect repair or optimization data selected by the generative large language model, a system performance optimization scheme is generated. The generative large language model includes a code generation model based on the Transformer architecture. The system performance optimization scheme is analyzed, and the target system is optimized.
[0006] Secondly, embodiments of this application also provide an AI-based performance analysis and optimization device, which adopts the following technical solution: AI-based performance analysis and optimization devices include: The data acquisition module is used to collect multi-source heterogeneous performance analysis data from the target system; The data analysis module is used to identify system performance defects by analyzing the performance analysis data; An optimization scheme generation module is used to input the system performance defects into a generative large language model, and generate a system performance optimization scheme based on the system performance defect repair or optimization data selected by the generative large language model. The generative large language model includes a code generation model based on the Transformer architecture. The performance optimization execution module is used to parse the system performance optimization scheme and optimize the performance of the target system.
[0007] Thirdly, embodiments of this application also provide a computer device that adopts the technical solution described below: A computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the AI-based performance analysis and optimization method described above.
[0008] Fourthly, embodiments of this application also provide a computer-readable storage medium, which adopts the technical solutions described below: A computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the steps of the AI-based performance analysis and optimization method described above.
[0009] Compared with the prior art, the embodiments of this application have the following main advantages: The AI-based performance analysis and optimization method described in this application collects multi-source heterogeneous performance analysis data from a target system; analyzes the performance analysis data to identify system performance defects; inputs the system performance defects into a generative large language model; generates a system performance optimization scheme based on the system performance defect repair or optimization data selected by the generative large language model; and parses the system performance optimization scheme to optimize the performance of the target system. By fully utilizing AI processing components in the data collection, data analysis, and subsequent system performance optimization scheme generation processes, significant manpower is saved, manual analysis steps are reduced, and the accuracy of the analysis results is ensured as much as possible. Applying this AI-based performance analysis and optimization method to system performance data analysis, especially in financial business scenarios with large and complex business data, can significantly improve analysis efficiency and promptly detect business performance anomalies. Attached Figure Description
[0010] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0011] Figure 1 This is an exemplary system architecture diagram to which this application can be applied; Figure 2 This is a flowchart of an embodiment of the AI-based performance analysis and optimization method according to this application; Figure 3 yes Figure 2 A flowchart of a specific embodiment of step 202 shown; Figure 4 yes Figure 3 A flowchart of a specific embodiment of step 301 shown; Figure 5 yes Figure 3 A flowchart of a specific embodiment of step 302 shown; Figure 6 yes Figure 2 A flowchart of a specific embodiment of step 203 shown; Figure 7 This is a schematic diagram of a structure of an embodiment of the AI-based performance analysis and optimization device according to this application; Figure 8 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation
[0012] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.
[0013] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0014] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.
[0015] like Figure 1 As shown, system architecture 100 may include terminal device 101, network 102, and server 103. Terminal device 101 may be a laptop 1011, tablet 1012, or mobile phone 1013. Network 102 is used as a medium to provide a communication link between terminal device 101 and server 103. Network 102 may include various connection types, such as wired, wireless communication links, or fiber optic cables.
[0016] Users can use terminal device 101 to interact with server 103 via network 102 to receive or send messages, etc. Various communication client applications can be installed on terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.
[0017] Terminal device 101 can be various electronic devices with a display screen and support web browsing. In addition to laptops 1011, tablets 1012, or mobile phones 1013, terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 player (Moving Picture Experts Group Audio Layer IV), a laptop computer, and a desktop computer, etc.
[0018] Server 103 can be a server that provides various services, such as a backend server that provides support for the pages displayed on terminal device 101.
[0019] It should be noted that the AI-based performance analysis and optimization method provided in this application embodiment is generally executed by a server, and correspondingly, the AI-based performance analysis and optimization device is generally set in the server.
[0020] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.
[0021] Continue to refer to Figure 2 The diagram illustrates a flowchart of an embodiment of the AI-based performance analysis and optimization method according to this application. The AI-based performance analysis and optimization method includes the following steps: Step 201: Collect multi-source heterogeneous performance analysis data from the target system.
[0022] In this embodiment, the target system may be, for example, a target financial business system, or more specifically, an insurance business system, a claims business system, an audit business system, a transaction management system, etc.
[0023] More specifically, the aforementioned multi-source heterogeneous performance analysis data refers to multi-formatted performance analysis data from different sources and channels, such as server resource utilization (CPU, memory), database response time, application logs, and distributed tracing data of the target financial business system.
[0024] By collecting multi-source heterogeneous performance analysis data from the target system, the system performance of the target system can be analyzed from the perspective of different performance analysis data, thus avoiding the one-sidedness of system performance problems analyzed by data from a single source.
[0025] In this embodiment, an AI-automated data acquisition component can also be used to automatically acquire multi-source heterogeneous performance analysis data from the target system, thereby improving data acquisition efficiency.
[0026] Step 202: By analyzing the performance analysis data, system performance defects are identified.
[0027] In this embodiment, the performance analysis data is analyzed in conjunction with a preset AI (artificial intelligence) processing method to identify system performance defects.
[0028] Specifically, two data analysis models were used to analyze the performance data: first, an association analysis model was used to analyze the performance data to identify feature data that showed a strong correlation with the performance anomaly indicators; then, an analytical large language model was used to perform a reasoning operation on the relevant business source code for the feature data that showed a strong correlation with the performance anomaly indicators, thereby identifying system performance defects.
[0029] In this embodiment, when analyzing the performance analysis data, an association analysis method was first adopted. Then, an analytical large language model was used to perform a reasoning of the relevant business source code. Compared with manual analysis and reasoning, this improved the efficiency of data analysis and greatly enhanced the efficiency of locating performance bottlenecks.
[0030] Step 203: Input the system performance defects into the generative large language model, and generate a system performance optimization scheme based on the system performance defect repair or optimization data selected by the generative large language model.
[0031] In this embodiment, pre-collected system performance defect repair or optimization data, such as multiple repair or optimization strategies and multiple repair or optimization processing steps, can be used as generation guidance data and imported into a generative large language model. Then, the generative large language model is used to optimize the system performance defects and generate corresponding system performance optimization solutions. For example, the system performance optimization solution includes a description of the system performance defect, an analysis of the root cause of the defect, specific code modification schemes, and expected performance improvement effects. The system performance optimization solution can integrate the analysis process, the location of the evidence chain, and optimization suggestions into a comprehensive performance diagnostic report, which is then presented to the user in the form of a comprehensive performance diagnostic report.
[0032] The generative large language model includes a code-generating model based on the Transformer architecture.
[0033] Step 204: Analyze the system performance optimization scheme and optimize the performance of the target system.
[0034] Specifically, by analyzing the system performance optimization scheme, the specific code modification scheme contained therein is obtained; based on the specific code modification scheme, the target system is optimized for performance. The specific code modification scheme includes task allocation configuration scheme, service resource balancing and distribution scheme, etc.
[0035] Specifically, assuming the target system is an insurance business processing system, the process involves collecting multi-source heterogeneous performance analysis data from the system; analyzing this data to identify performance defects; inputting these defects into a generative large language model; and generating performance optimization schemes based on data selected by the model for defect repair or optimization. Finally, these optimization schemes are analyzed to optimize the system's performance. By fully utilizing AI processing components in the data collection, analysis, and subsequent performance optimization scheme generation processes, significant manpower is saved, manual analysis steps are reduced, and the accuracy of the analysis results is maximized.
[0036] In this embodiment, multi-source heterogeneous performance analysis data is collected from the target system; system performance defects are identified by analyzing the performance analysis data; these defects are input into a generative large language model, and a system performance optimization plan is generated based on the system performance defect repair or optimization information selected by the generative large language model; the system performance optimization plan is then parsed to optimize the performance of the target system. By fully utilizing AI processing components in the data collection, data analysis, and subsequent system performance optimization plan generation processes, significant manpower consumption is saved, manual analysis steps are reduced, and the accuracy of the analysis results is ensured as much as possible. Applying this AI-based performance analysis and optimization method to system performance data analysis, especially in financial business scenarios with large and complex business data, can significantly improve analysis efficiency and promptly detect business performance anomalies.
[0037] In this embodiment, the step of collecting multi-source heterogeneous performance analysis data from the target system includes: collecting the performance analysis data from the target system using a data tracking method, wherein the multi-source heterogeneous performance analysis data includes application log data, system monitoring indicator data, link logic tracing data, and the target source code.
[0038] In this embodiment, after performing the step of collecting multi-source heterogeneous performance analysis data from the target system, the method further includes: preprocessing the performance analysis data, wherein the preprocessing includes data format unification conversion and data cleaning.
[0039] Specifically, since the collected performance analysis data is multi-source and heterogeneous, it may involve multiple databases or data in various formats. Therefore, after the performance analysis data is collected, a preset preprocessing component can be used to uniformly convert the data format of the performance analysis data to facilitate subsequent data analysis.
[0040] Continue to refer to Figure 3 , Figure 3 yes Figure 2 A flowchart of a specific embodiment of step 202 shown includes: Step 301: Perform a unified time-series feature vector transformation on the performance analysis data; Specifically, to facilitate subsequent AI processing, feature engineering is used to perform feature vectorization on the performance analysis data to obtain a unified time-series feature vector.
[0041] Step 302: By performing correlation analysis on the unified time-series feature vector, determine the target execution unit or target execution function when the performance index is abnormal; Specifically, by combining correlation analysis, the target execution unit or target execution function when performance indicators are abnormal can be determined. For example, statistical models such as Pearson correlation coefficient and Granger causality test can be used, or more complex time series analysis models based on Transformer can be used. When an abnormality is detected in a key business indicator, other features within the same time window will be analyzed immediately, and the correlation scores between them will be calculated to identify strong correlation features. Based on the identified strong correlation features, the target execution unit or target execution function when performance indicators are abnormal can be determined.
[0042] Step 303: Identify system performance defects based on the target execution unit or target execution function.
[0043] Continue to refer to Figure 4 , Figure 4 yes Figure 3 A flowchart of a specific embodiment of step 301 shown includes: Step 401: Based on the collection path of the performance analysis data, establish the association mapping relationship between the performance analysis data and the target source code file and the target code line; Specifically, since the collected performance analysis data includes application log data, system monitoring indicator data, link logic tracing data, and the target source code, the association mapping relationship between application log data, system monitoring indicator data, link logic tracing data, and the target source code can be constructed according to the collection path of the performance analysis data. Of course, the association mapping relationship between application log data, system monitoring indicator data, link logic tracing data, and the target code line can be established in a more granular way.
[0044] Assuming that the application log data records the execution log content related to the first object, the system monitoring indicator data also contains monitoring data for the first object, and the target source code contains the generation and calling functions of the first object, the application log data, system monitoring indicator data and the target source code can be associated with the first object as the association point.
[0045] Step 402: Based on the association mapping relationship and the execution timestamp corresponding to the target source code, the performance analysis data is processed by execution logic alignment and normalization transformation to construct a unified temporal feature vector containing system performance status, business application behavior and code context.
[0046] Specifically, the execution logic alignment and normalization transformation of the performance analysis data to construct a unified temporal feature vector containing system performance status, business application behavior, and code context essentially involves aligning application log data, system monitoring indicator data, and link logic tracing data with execution actions based on the execution time of the target source code. This achieves multi-dimensional alignment and constructs a unified temporal feature vector encompassing system performance status, business application behavior, and code context. This allows subsequent AI analysis components to fully utilize the alignment information and the unified temporal feature vector for system performance analysis.
[0047] Continue to refer to Figure 5 , Figure 5 yes Figure 3 A flowchart of a specific embodiment of step 302 shown includes: Step 501: Input the unified time series feature vector into the pre-trained association analysis model to identify strong correlation features used to characterize performance index anomalies. The pre-trained association analysis model includes an association analysis model based on Pearson correlation coefficient. The strong correlation features refer to features whose correlation exceeds a preset correlation threshold. Specifically, the pre-trained association analysis model can be obtained through supervised learning algorithms, based on historical performance monitoring data and its confirmed root cause labels. During training, a labeled historical performance monitoring dataset can be used. More specifically, the pre-trained association analysis model can incorporate statistical models such as Pearson correlation coefficient and Granger causality test, or more complex Transformer-based time series analysis models, for association analysis calculations.
[0048] Step 502: Based on the strong correlation features, filter out specific log errors, system resource bottleneck nodes and high latency nodes that cause abnormal performance indicators, wherein the high latency node refers to a node whose latency exceeds a preset latency threshold. Specifically, by using the strong correlation features, specific log errors, system resource bottleneck nodes, and high-latency nodes that cause abnormal performance indicators are screened out, so as to facilitate subsequent analysis of the system target source code by combining specific log errors, system resource bottleneck nodes, and high-latency nodes.
[0049] Step 503: Locate the target execution unit or target execution function through the specific log errors, system resource bottleneck nodes, and high latency nodes.
[0050] In this embodiment, the step of identifying system performance defects based on the target execution unit or target execution function specifically includes: performing static analysis on the source code involved in the target execution unit or target execution function using a Large Language Model (LLM) to identify system performance defects, wherein the Large Language Model includes a code analysis model based on the Transformer architecture.
[0051] Continue to refer to Figure 6 , Figure 6 yes Figure 2 A flowchart of a specific embodiment of step 203 shown includes: Step 601: Use the performance defect description field corresponding to the system performance defect as the optimization prompt word; In this embodiment, a generative large language model, which differs from the analytical large language model described above, is used. The two can have the same architecture, but differ in their processing mechanisms. The analytical large language model in the aforementioned steps focuses on analyzing and processing the input content, such as the source code involved in the target execution unit or target execution function; while the generative large language model focuses on generative processing of the input content, such as the performance defect description field corresponding to the system performance defect.
[0052] Step 602: Input the optimized prompt word into the generative large language model to obtain the Prompt template output by the generative large language model; Specifically, different Prompt templates are pre-imported into the generative large language model. The Prompt template can be understood as a template generated generatively based on the optimization prompt words. That is, the Prompt template is a pre-defined code structure used for effective data output. Generally, it includes three parts: input parameters, processing logic, and output results. The association and correspondence between different Prompt templates and different categories of optimization prompt words are pre-set.
[0053] Step 603: Select the step-by-step optimization steps with the highest confidence from the system performance defect repair or optimization data, and add them step by step to the Prompt template to generate the system performance optimization scheme.
[0054] Specifically, the step-by-step optimization process may include steps such as modifying the target source code, updating the source code, fixing bugs, and fixing vulnerabilities.
[0055] In this embodiment, multi-source heterogeneous performance analysis data is collected from the target system; system performance defects are identified by analyzing the performance analysis data; these defects are input into a generative large language model, and a system performance optimization plan is generated based on the system performance defect repair or optimization information selected by the generative large language model; the system performance optimization plan is then parsed to optimize the performance of the target system. By fully utilizing AI processing components in the data collection, data analysis, and subsequent system performance optimization plan generation processes, significant manpower consumption is saved, manual analysis steps are reduced, and the accuracy of the analysis results is ensured as much as possible. Applying this AI-based performance analysis and optimization method to system performance data analysis, especially in financial business scenarios with large and complex business data, can significantly improve analysis efficiency and promptly detect business performance anomalies.
[0056] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.
[0057] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.
[0058] In this embodiment, multi-source heterogeneous performance analysis data is collected from the target system; system performance defects are identified by analyzing the performance analysis data; these defects are input into a generative large language model, and a system performance optimization plan is generated based on the system performance defect repair or optimization information selected by the generative large language model; the system performance optimization plan is then parsed to optimize the performance of the target system. By fully utilizing AI processing components in the data collection, data analysis, and subsequent system performance optimization plan generation processes, significant manpower consumption is saved, manual analysis steps are reduced, and the accuracy of the analysis results is ensured as much as possible. Applying this AI-based performance analysis and optimization method to system performance data analysis, especially in financial business scenarios with large and complex business data, can significantly improve analysis efficiency and promptly detect business performance anomalies.
[0059] Further reference Figure 7 As a response to the above Figure 2 To implement the method shown, this application provides an embodiment of an AI-based performance analysis and optimization device, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.
[0060] like Figure 7 As shown, the AI-based performance analysis and optimization device 700 described in this embodiment includes: a data acquisition module 701, a data analysis module 702, an optimization scheme generation module 703, and a performance optimization execution module 704. Wherein: Data acquisition module 701 is used to acquire multi-source heterogeneous performance analysis data from the target system; Data analysis module 702 is used to identify system performance defects by analyzing the performance analysis data; The optimization scheme generation module 703 is used to input the system performance defects into the generative large language model, and generate a system performance optimization scheme based on the system performance defect repair or optimization data screened by the generative large language model. The generative large language model includes a code generation model based on the Transformer architecture. The performance optimization execution module 704 is used to parse the system performance optimization scheme and optimize the performance of the target system.
[0061] This application collects multi-source heterogeneous performance analysis data from a target system; analyzes the performance analysis data to identify system performance defects; inputs the system performance defects into a generative large language model; and generates a system performance optimization plan based on the system performance defect repair or optimization data selected by the generative large language model. The system performance optimization plan is then parsed to optimize the performance of the target system. By fully utilizing AI processing components in the data collection, data analysis, and subsequent system performance optimization plan generation processes, significant manpower is saved, manual analysis steps are reduced, and the accuracy of the analysis results is ensured as much as possible. Applying this AI-based performance analysis and optimization method to system performance data analysis, especially in financial business scenarios with large and complex business data, can significantly improve analysis efficiency and promptly detect business performance anomalies.
[0062] In this embodiment, the AI-based performance analysis and optimization device 700 further includes a data preprocessing module, which is used to preprocess the performance analysis data, wherein the preprocessing includes data format unification conversion and data cleaning.
[0063] In this embodiment, the data analysis module 702 includes a unified time-series feature vector conversion unit, a correlation analysis unit, and a system performance defect identification unit. Wherein: A unified time-series feature vector conversion unit is used to perform unified time-series feature vector conversion on the performance analysis data; The correlation analysis unit is used to determine the target execution unit or target execution function when the performance index is abnormal by performing correlation analysis on the unified time-series feature vector; The system performance defect identification unit is used to identify system performance defects based on the target execution unit or the target execution function.
[0064] In this embodiment, the unified temporal feature vector conversion unit is specifically used to establish an association mapping relationship between the performance analysis data and the target source code file and the target code line according to the collection path of the performance analysis data; and to perform execution logic alignment and normalization transformation on the performance analysis data according to the association mapping relationship and the execution timestamp corresponding to the target source code, so as to construct a unified temporal feature vector containing system performance status, business application behavior and code context.
[0065] In this embodiment, the correlation analysis unit is specifically used to input the unified time-series feature vector into a pre-trained correlation analysis model to identify strong correlation features used to characterize performance index anomalies. The pre-trained correlation analysis model includes a Pearson correlation coefficient-based correlation analysis model, and the strong correlation features refer to features whose correlation exceeds a preset correlation threshold. Based on the strong correlation features, specific log errors, system resource bottleneck nodes, and high-latency nodes that cause performance index anomalies are selected. The high-latency nodes refer to nodes whose latency exceeds a preset latency threshold. The target execution unit or target execution function is located through the specific log errors, system resource bottleneck nodes, and high-latency nodes.
[0066] In this embodiment, the system performance defect identification unit is specifically used to perform static analysis on the source code involved in the target execution unit or target execution function using an analytical large language model to identify system performance defects. The analytical large language model includes a code analysis model based on the Transformer architecture.
[0067] In this embodiment, the optimization scheme generation module 703 includes an optimization prompt word determination unit, a Prompt template selection unit, and an optimization scheme generation unit. Wherein: The optimization prompt word determination unit is used to use the performance defect description field corresponding to the system performance defect as the optimization prompt word; The Prompt template selection unit is used to input the optimized prompt words into the generative large language model and obtain the Prompt template output by the generative large language model. The optimization scheme generation unit is used to select the step-by-step optimization processing steps with the highest confidence from the system performance defect repair or optimization data, and add them step by step to the Prompt template to generate the system performance optimization scheme.
[0068] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).
[0069] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.
[0070] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 8 , Figure 8 This is a basic structural block diagram of the computer device in this embodiment.
[0071] The computer device 8 includes a memory 8a, a processor 8b, and a network interface 8c that are interconnected via a system bus. It should be noted that... Figure 8 Only a computer device 8 with component memory 8a, processor 8b, and network interface 8c is shown. However, it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Those skilled in the art will understand that the computer device described herein is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.
[0072] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.
[0073] The memory 8a includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 8a may be an internal storage unit of the computer device 8, such as the hard disk or memory of the computer device 8. In other embodiments, the memory 8a may also be an external storage device of the computer device 8, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. Of course, the memory 8a may include both internal storage units and external storage devices of the computer device 8. In this embodiment, the memory 8a is typically used to store the operating system and various application software installed on the computer device 8, such as computer-readable instructions for AI-based performance analysis and optimization methods. In addition, the memory 8a can also be used to temporarily store various types of data that have been output or will be output.
[0074] In some embodiments, the processor 8b may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 8b is typically used to control the overall operation of the computer device 8. In this embodiment, the processor 8b is used to execute computer-readable instructions stored in the memory 8a or to process data, for example, to execute the computer-readable instructions of the AI-based performance analysis and optimization method.
[0075] The network interface 8c may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 8 and other electronic devices.
[0076] The computer device proposed in this embodiment belongs to the field of artificial intelligence technology and is applied in system performance analysis and optimization scenarios. This application collects multi-source heterogeneous performance analysis data from a target system; identifies system performance defects by analyzing the performance analysis data; inputs the system performance defects into a generative large language model; generates a system performance optimization scheme based on the system performance defect repair or optimization data selected by the generative large language model; and parses the system performance optimization scheme to optimize the performance of the target system. By fully utilizing AI processing components in the data collection, data analysis, and subsequent system performance optimization scheme generation processes, significant manpower consumption is saved, manual analysis steps are reduced, and the accuracy of the analysis results is ensured as much as possible. Applying this AI-based performance analysis and optimization method to system performance data analysis, especially in financial business scenarios with large and complex business data, can significantly improve analysis efficiency and promptly detect business performance anomalies.
[0077] This application also provides another implementation, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by a processor to cause the processor to perform the steps of the AI-based performance analysis and optimization method described above.
[0078] The computer-readable storage medium proposed in this embodiment belongs to the field of artificial intelligence technology and is applied in system performance analysis and optimization scenarios. This application collects multi-source heterogeneous performance analysis data from a target system; identifies system performance defects by analyzing the performance analysis data; inputs the system performance defects into a generative large language model; generates a system performance optimization scheme based on the system performance defect repair or optimization data selected by the generative large language model; and parses the system performance optimization scheme to optimize the performance of the target system. By fully utilizing AI processing components in the data collection, data analysis, and subsequent system performance optimization scheme generation processes, significant manpower is saved, manual analysis steps are reduced, and the accuracy of the analysis results is ensured as much as possible. Applying this AI-based performance analysis and optimization method to system performance data analysis, especially in financial business scenarios with large and complex business data, can significantly improve analysis efficiency and promptly detect business performance anomalies.
[0079] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.
[0080] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to make the disclosure of this application more thorough and comprehensive. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application. Software tools or components not belonging to this company that appear in the embodiments of this application are merely illustrative examples and do not represent actual use.
Claims
1. An AI-based performance analysis and optimization method, characterized in that, Includes the following steps: Collect multi-source heterogeneous performance analysis data from the target system; By analyzing the performance analysis data, system performance defects are identified; The system performance defects are input into the generative large language model, and a system performance optimization scheme is generated based on the system performance defect repair or optimization data selected by the generative large language model. The system performance optimization scheme is analyzed, and the target system is optimized.
2. The AI-based performance analysis and optimization method according to claim 1, characterized in that, The steps for collecting multi-source heterogeneous performance analysis data from the target system include: The performance analysis data is collected from the target system using a data tracking method. The multi-source heterogeneous performance analysis data includes application log data, system monitoring indicator data, and link logic tracing data. After performing the step of collecting multi-source heterogeneous performance analysis data from the target system, the method further includes: The performance analysis data is preprocessed, including data format unification conversion and data cleaning.
3. The AI-based performance analysis and optimization method according to claim 1 or 2, characterized in that, The step of identifying system performance defects by analyzing the performance analysis data specifically includes: The performance analysis data is subjected to a unified time-series feature vector transformation; By performing correlation analysis on the unified temporal feature vector, the target execution unit or target execution function when the performance index is abnormal can be determined. Based on the target execution unit or target execution function, identify system performance defects.
4. The AI-based performance analysis and optimization method according to claim 3, characterized in that, The step of performing a unified time-series feature vector transformation on the performance analysis data specifically includes: Based on the collection path of the performance analysis data, establish an association mapping relationship between the performance analysis data and the target source code file and the target code line; Based on the association mapping relationship and the execution timestamp corresponding to the target source code, the performance analysis data is processed by execution logic alignment and normalization transformation to construct a unified temporal feature vector containing system performance status, business application behavior and code context.
5. The AI-based performance analysis and optimization method according to claim 3, characterized in that, The step of determining the target execution unit or target execution function when performance indicators are abnormal by performing correlation analysis on the unified time-series feature vector specifically includes: The unified time-series feature vector is input into a pre-trained association analysis model to identify strong correlation features used to characterize performance index anomalies. The pre-trained association analysis model includes an association analysis model based on Pearson correlation coefficient, and the strong correlation features refer to features whose correlation exceeds a preset correlation threshold. Based on the strong correlation characteristics, specific log errors, system resource bottleneck nodes, and high-latency nodes that cause abnormal performance indicators are screened out. The high-latency nodes refer to nodes whose latency exceeds a preset latency threshold. The target execution unit or target execution function can be located by identifying specific log errors, system resource bottleneck nodes, and high-latency nodes.
6. The AI-based performance analysis and optimization method according to claim 3, characterized in that, The step of identifying system performance defects based on the target execution unit or target execution function specifically includes: Static analysis of the source code involved in the target execution unit or target execution function is performed using an analytical large language model to identify system performance defects.
7. The AI-based performance analysis and optimization method according to claim 6, characterized in that, The step of inputting the system performance defects into a generative large language model, and generating a system performance optimization scheme based on the system performance defect repair or optimization data selected by the generative large language model, specifically includes: The performance defect description field corresponding to the system performance defect is used as the optimization prompt word; The optimized prompt words are input into the generative large language model to obtain the Prompt template output by the generative large language model; From the system performance defect repair or optimization data, the step-by-step optimization steps with the highest confidence are selected and added to the Prompt template step by step to generate the system performance optimization plan.
8. An AI-based performance analysis and optimization device, characterized in that, include: The data acquisition module is used to collect multi-source heterogeneous performance analysis data from the target system; The data analysis module is used to identify system performance defects by analyzing the performance analysis data; The optimization scheme generation module is used to input the system performance defects into the generative large language model, and generate system performance optimization schemes based on the system performance defect repair or optimization data selected by the generative large language model. The performance optimization execution module is used to parse the system performance optimization scheme and optimize the performance of the target system.
9. A computer device, characterized in that, The system includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the AI-based performance analysis and optimization method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the AI-based performance analysis and optimization method as described in any one of claims 1 to 7.