Processing capability evaluation method and device

Abstract

The invention discloses a processing capability evaluation method and device and relates to the field of communication. On the premise that evaluation cost is reduced, processing capability evaluation of a large-scale computer cluster is achieved. According to the specific scheme, operation parameters of a small-scale computer cluster during operation and application are collected, a performance deduction model is obtained according to the operation parameters, configuration information of the small-scale computer cluster and configuration information of the large-scale computer cluster are obtained, and the processing capability of the large-scale computer cluster is evaluated according to the configuration information of the small-scale computer cluster, the configuration information of the large-scale computer cluster and the performance deduction model to obtain the processing capability of the large-scale computer cluster. The method and device are used in the processing capability evaluation process of the large-scale computer cluster.

Description

technical field [0001] The present invention relates to the communication field, in particular to a processing capability evaluation method and device. Background technique [0002] With the development of science and technology and the Internet, the amount of information in modern society is increasing rapidly, and large-scale data is accumulated in this information, and some of these data will be stored in the cloud platform or processed with the help of the cloud platform. With the help of Hadoop, users can write distributed parallel programs and run them on large-scale computer clusters consisting of hundreds or thousands of nodes to efficiently store, manage and analyze without knowing the underlying details of the distribution. These data are stored in the cloud platform. [0003] Hadoop is an open source software architecture capable of distributed processing of large amounts of data. At the bottom is a distributed file system, which uses distributed storage to store...

AI Technical Summary

Problems solved by technology

[0004] First of all, when evaluating large-scale computer clusters, applications need to be deployed on large-scale computer clusters, but due to the large scale of computer clusters (hundreds or even thousands of nodes), it is difficult to find a large-scale computer cluster for evaluation. Large-scale computer clusters. In addition, if the evaluation is performed directly on the production system, it will affect the normal operation of the production system and even bring disasters to the production system. Therefore, it is impossible to find a suitable large-scale computer cluster for system-wide processing capability evaluation.

[0005] Secondly, even if a suitable large-scale computer cluster is found, it will take a long time and consume a lot of energy to evaluate the processing power of the whole system on the computer cluster consisting of hundreds or even thousands of nodes, resulting in Increased assessment costs

Images