Collaborative analysis and localization method for multi-source faults for large-scale systems

Abstract

The invention discloses a large-scale system-oriented multi-source fault collaborative analysis and positioning method, which includes the following steps: S1. Unified classification of faults collected by each fault monitoring module, defining a fault code Fid for each fault, and assigning a fault code Fid to each fault Fault definition upper and lower association lists Fuplist and Fdownlist, the upper association list Fuplist contains a group of fault codes Fid that will induce the fault, and the lower association list Fdownlist contains a group of fault Fids that the fault will induce; S2. The fault analysis system receives information from each fault The faults sent by the monitoring module form a current fault list; S3, the fault analysis system performs contextual analysis on the current fault list; S10, the fault analysis system realizes the precise positioning of a fault Fk, and jumps to S4. The invention improves the ability of automatic analysis and positioning of system faults, and solves the problem of accurate positioning of large-scale parallel system faults.

Description

technical field [0001] The invention relates to a large-scale system-oriented multi-source fault collaborative analysis and positioning method, belonging to the technical field of availability management of large-scale parallel systems. Background technique [0002] With the continuous development of high-performance computing technology, the scale of large-scale parallel systems is increasing day by day, the resources are diverse and large in quantity, the structure of software and hardware is complex, and the number of components is increasing, resulting in the mean time between failures (MTBF) of large-scale systems. Significantly reduced, component failure has become a normal event in the process of system operation, and the fault-tolerant system has become the basic support system for the stable and reliable operation of the system. Rapid discovery and accurate location of system faults are the basis for system fault handling and fault-tolerant operation. [0003] In m...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a large-scale system-oriented multi-source fault collaborative analysis and positioning method, which improves the system fault automatic analysis and positioning ability, and solves the problem of large-scale parallel system faults The problem of accurate positioning

Images