Method and device for performance detection and fault location based on operating system, and medium

Abstract

The invention relates to the technical field of system fault detection, and discloses an operating system-based performance detection and fault positioning method, which comprises the following steps of: S1, performance detection: judging whether a current system is in a bottleneck state, judging which layer of a multi-layer architecture the bottleneck appears in, and judging which resources the bottleneck is caused by, if yes, judging whether the current system is in the bottleneck state; the performance of the operating system is detected and evaluated; s2, fault positioning: positioning a fault, analyzing a source and a reason of the fault and processing the fault in a mode of combining static analysis and fault tree analysis; and S3, fault classification: classifying the processed faults according to types, and establishing a fault detection model to realize early warning and fault prevention. The method comprises the following steps: performing performance detection and evaluation on an operating system; positioning the fault, analyzing the source and reason of the fault, and processing the fault; and finally, establishing a fault detection model to realize early warning and fault prevention of the fault of the operating system.

Description

Technical Field

[0001] The present invention relates to the technical field of system fault detection, and in particular to a method, device and medium based on operating system performance detection and fault location. Background Art

[0002] With the continuous progress of society and the rapid development of information technology, the application of computers has penetrated into all aspects of production and life in modern society, realizing the informatization, intelligence and automation of related work. People's demand and dependence on computer systems are increasing; and the operating system is an important part of the computer and the soul of the computer; among them, large-scale and complex operating systems represented by large enterprise application systems and cloud computing are becoming increasingly widespread, and performance tuning, performance management and capacity planning have become important challenges faced in the deployment and operation of such operating systems.

[0003] A Chinese patent discloses a troubleshooting method, apparatus, computer equipment and storage medium (Announcement No. CN118331819A). The patented technology obtains the version information of the operating system, in which the environment deployment data of the application is stored; determines the command type corresponding to the version information, and determines the troubleshooting result of the environment deployment data according to the detection command corresponding to the command type; generates the fault information of the environment deployment based on the troubleshooting result; can reduce the missed fault detection rate of the environment deployment to improve the safety performance of the entire operating system, but it cannot detect and evaluate the performance of the operating system, and cannot find the source and cause of the fault, which leads to the inability to effectively warn before the fault occurs. Summary of the invention

[0004] The purpose of the present invention is to provide a method, device and medium based on operating system performance detection and fault location to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: The method based on operating system performance detection and fault location includes the following steps: S1. Performance testing: Use multi-threading to simulate multiple users accessing the operating system server, and collect various performance data through the data collection module; based on the collected performance data, continuously increase the system load pressure, and use the performance evaluation module to determine whether the current system is in a bottleneck state. If it is in a bottleneck state, determine which layer of the multi-layer architecture the bottleneck occurs in, and analyze and determine which resources cause the bottleneck, so as to achieve performance testing and evaluation of the operating system; S2. Fault location: Arrange the data collected by the data acquisition module in ascending order at certain time intervals, and then use the fault location module to locate the fault and analyze the source and cause of the fault by combining static analysis and fault tree analysis; and use the fault handling module to handle the impact of the fault on the operating system's working status, stability, and key service reliability; S3. Fault classification: Based on the dynamic fault detection strategy, a fault classification module is used to classify the processed faults according to their types, and a fault detection model is established. The fault detection model is used to monitor the operating status and key parameters of the operating system in real time, and the probability of system failure is predicted, so that corresponding measures can be taken in advance to achieve early warning and fault prevention.

[0006] As a further solution of the present invention: in the S1 step, the performance data collected by the data collection module includes: the average response time of the server, the shortest and longest response time of the server, the service data completed by the server per second, the server call success rate, and the total interaction data and total running time of the server.

[0007] As a further solution of the present invention: in the step S2, the specific implementation method of combining static analysis and fault tree analysis is as follows: S21, statically analyzing the operating system program statements, combining the information of lexical analysis and syntax analysis to complete the analysis of the source code, find out various faults, and locate them; S22. Next, a fault tree is constructed using the found fault source code, and then the minimum cut set that causes the top event is calculated using Boolean algebra, and the source and cause of the fault are found based on the minimum cut set.

[0008] As a further solution of the present invention: in the step S2, the fault processing module includes a fault diagnosis unit, a fault analysis unit, a fault execution unit and a fault learning unit, wherein: The fault diagnosis unit is used to determine the type of fault, the source of the fault, and the possible impact range of the fault based on the acquired fault status and behavior information; The fault analysis unit is used to infer the development trend of the fault state based on the fault diagnosis structure and combined with the fault knowledge base information, so as to detect larger system faults or system failures in the future; The fault execution unit is used to plan the fault handling behavior to be performed based on the analysis and diagnosis results and using the fault rule library; and execute the corresponding instructions to complete the fault handling; The fault learning sheet is used to analyze the fault handling process and operating system fault characteristics, form new fault knowledge or rules, and add them to the fault knowledge base and fault rule base.

[0009] As a further solution of the present invention: in the S3 step, the classified fault types include application layer faults, kernel layer faults and hardware layer faults; wherein, application layer faults include user configuration service class faults, system key service class faults and system core service class faults, and kernel layer faults include kernel numerical faults and kernel non-numerical faults.

[0010] As a further solution of the present invention: in the step S3, the fault detection model is composed of a fault detection tool, a fault detection object, a fault detection algorithm, a historical information storage unit and a fault database, wherein: Fault detection tools are used to detect fault types, including user configuration service detection tools, system key service detection tools, system core service detection tools, kernel numerical fault detection tools, kernel non-numerical fault detection tools, and hardware layer fault detection tools; The fault detection object is used to divide the classified faults into multiple fault units; The fault detection algorithm uses a combination of static analysis and fault tree analysis; The historical information storage unit includes a fault monitoring result file and a fault record file, which are used to store periodic fault information and detected fault information, wherein each fault detection object corresponds to a fault monitoring result file, and each fault detection tool corresponds to a fault record file; The fault database stores all fault detection tools, fault detection objects, fault detection algorithms and historical information storage units.

[0011] The device based on the method of operating system performance detection and fault location includes a data acquisition module, a performance evaluation module, a fault location module, a fault processing module and a fault classification module, wherein: The data collection module is used to collect various performance data of the server; The performance evaluation module is used to detect and evaluate the performance indicators of the operating system; The fault location module is used to locate the fault and analyze the source and cause of the fault; The fault handling module is used to handle the impact of faults on the operating system's working status, stability, and key service reliability; The fault classification module classifies the processed faults according to their types and establishes a fault detection model to achieve early warning and fault prevention of the operating system.

[0012] A medium based on a method for operating system performance detection and fault location, wherein a computer operating system program is stored on the medium and is used to implement operating system performance detection and fault location.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The present invention first collects various performance data of the operating system server and determines the resources that cause system bottlenecks to achieve performance detection and evaluation of the operating system; then uses a combination of static analysis and fault tree analysis to locate the fault, analyze the source and cause of the fault, and handle the fault; then, classify the handled faults according to type, and establish a fault detection model to provide early warning and fault prevention for operating system faults. DETAILED DESCRIPTION

[0014] In an embodiment of the present invention, a method for operating system performance detection and fault location includes the following steps: S1. Performance testing: Use multi-threading to simulate multi-user access to the operating system server, and collect various performance data through the data collection module; based on the collected performance data, continuously increase the system load pressure, and use the performance evaluation module to determine whether the current system is in a bottleneck state. If the system throughput drops sharply, then determine whether the current system is in a bottleneck state; if it is in a bottleneck state, then determine which layer of the multi-layer architecture the bottleneck appears in. The characteristic of the bottleneck layer is that as the load pressure continues to increase, its execution time increases the fastest, so as to find out which layer the bottleneck is in; and analyze and determine which resources cause the bottleneck, such as high usage, and the growth rate gradually becomes 0; to achieve performance testing and evaluation of the operating system; S2. Fault location: Arrange the data collected by the data acquisition module in ascending order at certain time intervals, and then use the fault location module to locate the fault and analyze the source and cause of the fault by combining static analysis and fault tree analysis; and use the fault handling module to handle the impact of the fault on the operating system's working status, stability, and key service reliability; S3. Fault classification: Based on the dynamic fault detection strategy, a fault classification module is used to classify the processed faults according to their types, and a fault detection model is established. The fault detection model is used to monitor the operating status and key parameters of the operating system in real time, and the probability of system failure is predicted, so that corresponding measures can be taken in advance to achieve early warning and fault prevention.

[0015] Preferably, in the S1 step, the performance data collected by the data collection module include: average response time of the server, shortest and longest response time of the server, service data completed by the server per second, server call success rate, and total interaction data and total running time of the server.

[0016] Preferably, in the step S2, the specific implementation method of combining static analysis and fault tree analysis is as follows: S21, statically analyzing the operating system program statements, combining the information of lexical analysis and syntax analysis to complete the analysis of the source code, find out various faults, and locate them; S22. Next, a fault tree is constructed using the found fault source code, and then the minimum cut set that causes the top event is calculated using Boolean algebra, and the source and cause of the fault are found based on the minimum cut set.

[0017] Preferably, in the step S2, the fault processing module includes a fault diagnosis unit, a fault analysis unit, a fault execution unit and a fault learning unit, wherein: The fault diagnosis unit is used to determine the type of fault, the source of the fault, and the possible impact range of the fault based on the acquired fault status and behavior information; The fault analysis unit is used to infer the development trend of the fault state based on the fault diagnosis structure and combined with the fault knowledge base information, so as to detect larger system faults or system failures in the future; The fault execution unit is used to plan the fault handling behavior to be performed based on the analysis and diagnosis results and using the fault rule library; and execute the corresponding instructions to complete the fault handling; The fault learning sheet is used to analyze the fault handling process and operating system fault characteristics, form new fault knowledge or rules, and add them to the fault knowledge base and fault rule base.

[0018] Preferably, in the S3 step, the classified fault types include application layer faults, kernel layer faults and hardware layer faults; wherein, application layer faults include user configuration service class faults, system key service class faults and system core service class faults, and kernel layer faults include kernel numerical faults and kernel non-numerical faults.

[0019] Preferably, in the step S3, the fault detection model is composed of a fault detection tool, a fault detection object, a fault detection algorithm, a historical information storage unit and a fault database, wherein: Fault detection tools are used to detect fault types, including user configuration service detection tools, system key service detection tools, system core service detection tools, kernel numerical fault detection tools, kernel non-numerical fault detection tools, and hardware layer fault detection tools; The fault detection object is used to divide the classified faults into multiple fault units; The fault detection algorithm uses a combination of static analysis and fault tree analysis; The historical information storage unit includes a fault monitoring result file and a fault record file, which are used to store periodic fault information and detected fault information, wherein each fault detection object corresponds to a fault monitoring result file, and each fault detection tool corresponds to a fault record file; The fault database stores all fault detection tools, fault detection objects, fault detection algorithms and historical information storage units.

[0020] The device based on the method of operating system performance detection and fault location includes a data acquisition module, a performance evaluation module, a fault location module, a fault processing module and a fault classification module, wherein: The data collection module is used to collect various performance data of the server; The performance evaluation module is used to detect and evaluate the performance indicators of the operating system; The fault location module is used to locate the fault and analyze the source and cause of the fault; The fault handling module is used to handle the impact of faults on the operating system's working status, stability, and key service reliability; The fault classification module classifies the processed faults according to their types and establishes a fault detection model to achieve early warning and fault prevention of the operating system.

[0021] A medium based on a method for operating system performance detection and fault location, wherein a computer operating system program is stored on the medium and is used to implement operating system performance detection and fault location.

[0022] What is described above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method based on operating system performance detection and fault location, characterized in that: The following steps are involved: S1. Performance testing: simulate multiple users accessing the operating system server in a multi-threaded manner, and collect various performance data through the data collection module; Based on the collected performance data, the system load pressure is continuously increased, and the performance evaluation module is used to determine whether the current system is in a bottleneck state. If it is in a bottleneck state, it is determined at which layer of the multi-layer architecture the bottleneck occurs, and the bottleneck is analyzed and determined by which resources, so as to achieve performance detection and evaluation of the operating system. S2. Fault location: Arrange the data collected by the data acquisition module in ascending order at certain time intervals, and then use the fault location module to locate the fault and analyze the source and cause of the fault by combining static analysis and fault tree analysis; and use the fault handling module to handle the impact of the fault on the operating system's working status, stability, and key service reliability; S3. Fault classification: Based on the dynamic fault detection strategy, a fault classification module is used to classify the processed faults according to their types, and a fault detection model is established. The fault detection model is used to monitor the operating status and key parameters of the operating system in real time, and the probability of system failure is predicted, so that corresponding measures can be taken in advance to achieve early warning and fault prevention.

2. The method for operating system performance detection and fault location according to claim 1, characterized in that: In the step S1, the performance data collected by the data collection module include: average response time of the server, shortest and longest response time of the server, service data completed by the server per second, server call success rate, and total interaction data and total running time of the server.

3. The method for operating system performance detection and fault location according to claim 1, characterized in that: In the step S2, the specific implementation method of combining static analysis and fault tree analysis is as follows: S21, statically analyzing the operating system program statements, combining the information of lexical analysis and syntax analysis to complete the analysis of the source code, find out various faults, and locate them; S22. Next, a fault tree is constructed using the found fault source code, and then the minimum cut set that causes the top event is calculated using Boolean algebra, and the source and cause of the fault are found based on the minimum cut set.

4. The method for operating system performance detection and fault location according to claim 1, characterized in that: In the step S2, the fault processing module includes a fault diagnosis unit, a fault analysis unit, a fault execution unit and a fault learning unit, wherein: The fault diagnosis unit is used to determine the type of fault, the source of the fault, and the possible impact range of the fault based on the acquired fault status and behavior information; The fault analysis unit is used to infer the development trend of the fault state based on the fault diagnosis structure and combined with the fault knowledge base information, so as to detect larger system faults or system failures in the future; The fault execution unit is used to plan the fault handling behavior to be performed based on the analysis and diagnosis results and using the fault rule library; and execute the corresponding instructions to complete the fault handling; The fault learning sheet is used to analyze the fault handling process and operating system fault characteristics, form new fault knowledge or rules, and add them to the fault knowledge base and fault rule base.

5. The method based on operating system performance detection and fault location according to claim 1, characterized in that: In the S3 step, the classified fault types include application layer faults, kernel layer faults and hardware layer faults; wherein, application layer faults include user configuration service class faults, system key service class faults and system core service class faults, and kernel layer faults include kernel numerical faults and kernel non-numerical faults.

6. The method for operating system performance detection and fault location according to claim 1, characterized in that: In the step S3, the fault detection model is composed of a fault detection tool, a fault detection object, a fault detection algorithm, a historical information storage unit and a fault database, wherein: Fault detection tools are used to detect fault types, including user configuration service detection tools, system key service detection tools, system core service detection tools, kernel numerical fault detection tools, kernel non-numerical fault detection tools, and hardware layer fault detection tools; The fault detection object is used to divide the classified faults into multiple fault units; The fault detection algorithm uses a combination of static analysis and fault tree analysis; The historical information storage unit includes a fault monitoring result file and a fault record file, which are used to store periodic fault information and detected fault information, wherein each fault detection object corresponds to a fault monitoring result file, and each fault detection tool corresponds to a fault record file; The fault database stores all fault detection tools, fault detection objects, fault detection algorithms and historical information storage units.

7. A device for implementing the method of operating system performance detection and fault location according to claim 1, characterized in that: It includes data acquisition module, performance evaluation module, fault location module, fault processing module and fault classification module, among which: The data collection module is used to collect various performance data of the server; The performance evaluation module is used to detect and evaluate the performance indicators of the operating system; The fault location module is used to locate the fault and analyze the source and cause of the fault; The fault handling module is used to handle the impact of faults on the operating system's working status, stability, and key service reliability; The fault classification module classifies the processed faults according to their types and establishes a fault detection model to achieve early warning and fault prevention of the operating system.

8. A medium for implementing the method for operating system performance detection and fault location according to claim 1, characterized in that: The medium stores a computer operating system program for implementing performance detection and fault location of the operating system.