Fault statistics method orienting HPC cluster and fault statistics device thereof

Abstract

The invention provides a fault statistics method orienting an HPC cluster and a fault statistics device thereof. The fault statistics method comprises a fault write-in process and a fault display process. The fault write-in process comprises the steps that the fault data of multiple servers are inputted, wherein the fault data of multiple servers are inputted through a form input mode; and the fault data of multiple servers are acquired, and the fault data of multiple servers are stored in a MySQL database through a PHP script. The fault display process comprises the steps that fault query information is acquired; and the fault data stored in the MySQL database are displayed according to the fault query information, wherein the fault data stored in the MySQL database are displayed through an HTML form. The data are written in the database by using the PHP and the HTML form based on the MySQL database, and the data are displayed to the user through the HTML form so that time and error of manual operation can be reduced and query and statistics are easy.

Description

technical field [0001] The invention relates to the field of computers, in particular to a fault statistics method and device for HPC clusters. Background technique [0002] In the link of HPC (High Performance Computing, high-performance computing) order joint debugging, that is, after the server completes the aging test, the user-customized servers of various models are uniformly installed in the application environment for testing. At the same time, some faults will inevitably be found during the test. However, the existing technology is to manually count the fault data into the data in the excel table. However, the above-mentioned method has defects such as easy to generate errors and untimely statistics. At the same time, it is manually completed a series of operations, which consumes Manpower, inefficient, and error-prone. [0003] Aiming at the problems in the related technologies, no effective solution has been proposed yet. Contents of the invention [0004] Aim...

AI Technical Summary

Problems solved by technology

[0002] In the link of HPC (High Performance Computing, high-performance computing) order joint debugging, that is, after the server completes the aging test, the user-customized servers of various models are uniformly installed in the application environment for testing. At the same time, some faults will inevitably be found during the test. However, the existing technology is to manually count the fault data into the data in the excel table. However, the above-mentioned method has defects such as easy to generate errors and untimely statistics. At the same time, it is manually completed a series of operations, which consumes Manpower, inefficient, and error-prone

Images