Unlock instant, AI-driven research and patent intelligence for your innovation.

Database cluster server data classification method

A cluster server and data classification technology, applied in the field of monitoring and statistics of key indicators of database cluster servers, can solve problems such as inability to facilitate database cluster server data invocation, affecting data processing efficiency of high-performance computers, and disadvantageous database data classification, etc. To achieve the effect of convenient graphical display, accurate and reliable analysis results, and simple and clear classification results

Active Publication Date: 2017-06-16
ZHENGZHOU XIZHI INFORMATION TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

It has the disadvantages of long calculation time and low processing efficiency, which is not conducive to quickly and efficiently classifying database data, is not conducive to realizing the monitoring and statistics of key indicators of database cluster servers, and cannot conveniently and quickly implement database cluster server The call of data affects the data processing efficiency of high-performance computers

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Database cluster server data classification method
  • Database cluster server data classification method
  • Database cluster server data classification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] The database cluster server data classification method of the present embodiment, its implementation steps are as follows:

[0027] 1) Configure a monitoring agent instance on each server in the database cluster server, and configure a global data server in the database cluster server to collect and display the data of each monitoring agent instance;

[0028] 2) Each monitoring agent instance obtains the LOAD, QPS, TPS, network I / O, disk I / O, memory usage, and CPU usage of the current server at regular intervals, summarizes them to the global data server, and establishes a database table ;

[0029] 3) The global data server classifies and summarizes the host name and time information according to the LOAD, QPS, and TPS indicators of each server, and obtains various monitoring data information of all hosts within a certain period of time;

[0030] 4) Combining the characteristics of the relational database, read the data once according to the time range when fetching da...

Embodiment 2

[0038] The database cluster server data classification method of this embodiment differs from Embodiment 1 in that in step 2) and step 3), each monitoring agent instance obtains the LOAD, QPS, and QPS of the current server every 2 to 5 seconds. TPS, network I / O, disk I / O, memory usage, CPU usage information; the structure of the database table established by the global data server includes fields and types, and the fields include host name, detection time, server load, disk per second The number of reads, the number of disk writes per second, the number of sectors read by the disk per second, the number of sectors written by the disk per second, the size of the disk IO queue, the ratio of disk waits, the number of Db insertions, the number of Db updates, and the number of Db deletions Number of times, Db query times, cache hit rate, and then establish a database table through the global data server to classify them by type.

Embodiment 3

[0040] In the database cluster server data classification method of this embodiment, the steps for realizing three-dimensional classification of database cluster server data and monitoring and statistics of key indicators are as follows:

[0041] 1. Configure a global data server for data collection and data display in the server cluster system; configure a monitoring agent instance on each server in the server cluster system;

[0042] 2. The monitoring agent instance obtains the LOAD, QPS, TPS, network I / O, disk I / O, memory usage, and CPU usage of the current server every 3 minutes, and summarizes them to the global data collection server, and the global data server builds a database table . The database table structure and description are as follows:

[0043] field Types of illustrate hostname varchar(50) CPU name datetime datetime detection time load_5m float server load io_r float Disk reads per second io_w float d...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a method for analyzing, processing and classifying data of a database cluster server. The method for classifying the data of the database cluster server comprises the following implementation steps that a global data server is arranged in the database cluster server; each server body is provided with a monitoring agency instance; data information of the current server body is obtained at intervals through the corresponding monitoring agency instance, and is collected into the global data server, and a database list is built; the global data server carries out classification collection according to LOAD indexes, OPS indexes and TPS indexes of the server bodies, host names and time information, and obtains different types of monitoring data information of all hosts in a certain period; data reading is carried out for one time according to the characteristics of a relational database and the combination of the time ranges, so that a data set A is obtained; combination between a one-time database reading operation and the large and small barrel method is adopted, small barrels with the same index portions in names of small barrels are placed into corresponding large index barrels, and therefore monitoring and statistics of key indexes of the database cluster server are achieved.

Description

technical field [0001] The invention relates to a data analysis, processing and classification method of a database cluster server, which is applied to the monitoring and statistics of key indicators of the database cluster server. Background technique [0002] A database cluster uses multiple database servers to form a virtual single database logical image, which, like a single database system, can provide transparent data services to clients. With the advent of the era of big data, the scale of database clusters is getting larger and larger, and the performance monitoring and analysis of the cluster system is particularly important, that is, it is necessary to have an intuitive understanding of the overall operation of the entire cluster, and to There is a fast, clear and accurate viewing and positioning of the operation status of each server. [0003] The traditional relational database is a two-dimensional database, through which the data classification of the database ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F17/30
CPCG06F16/285
Inventor 李国强王路涂新行
Owner ZHENGZHOU XIZHI INFORMATION TECH