Performance analysis method, system and equipment for non-structured distributed storage system

Abstract

The application discloses a performance analysis method, system and equipment for a non-structured distributed storage system. The method comprises the following steps: acquiring monitoring data monitored by a monitor running on a server side, wherein monitors running on different servers are subjected to a time synchronization operation before starting; acquiring performance data of a client; andassociating the performance data with the monitoring data on the same time axis, and analyzing the associated performance data and monitoring data in order to generate a corresponding performance analysis result. According to the performance analysis method disclosed by the invention, the acquired performance data and monitoring data are associated on the same reference time axis, so that deeperperformance analysis and rating of the non-structured distributed storage system are realized, an; a developer is assisted in finding and determining a potential optimizing point in order to realize the performance optimization of the non-structured distributed storage system.

Description

technical field [0001] The invention relates to the field of distributed storage, in particular to a performance analysis method, system and equipment of an unstructured distributed storage system. Background technique [0002] Unstructured distributed storage technology is produced for the application requirements of massive data storage, high concurrent reading and writing of data, and high scalability. It is different from the application scenarios of traditional relational database systems. Since the unstructured distributed storage system has the characteristics of massive data storage, high performance, high availability, and strong scalability, and is different from the application scenarios of traditional relational databases, the performance analysis tools of traditional relational databases cannot be directly used for Structured distributed storage system for performance analysis. In the prior art, the performance analysis of the unstructured distributed storage s...

AI Technical Summary

Problems solved by technology

Since the unstructured distributed storage system has the characteristics of massive data storage, high performance, high availability, and strong scalability, and is different from the application scenarios of traditional relational databases, the performance analysis tools of traditional relational databases cannot be directly used for Performance Analysis of Structured Distributed Storage System

In the prior art, the performance analysis of the unstructured distributed storage system is performed from the perspective of the client or the server. However, the former can only be used to help developers choose a suitable storage system for their own application scenarios, and cannot achieve Performance tuning of unstructured distributed storage systems; the latter is usually used to find optimization points or optimization methods, or to clarify analysis points, and then conduct targeted monitoring to verify these tuning points or analysis points The impact on system performance, so the performance tuning of unstructured distributed storage systems cannot be realized when the analysis point is not clear, or there is no clear optimization point and optimization path

Images