Distributed cluster performance optimization method and device, terminal and storage medium

A distributed cluster and optimization method technology, applied in the field of cluster performance optimization, can solve problems such as slow response, decreased read and write performance, and affect user experience, and achieve the effects of reducing impact, simple structure, and preventing the overall performance of the cluster from decreasing

Inactive Publication Date: 2021-01-08
INSPUR SUZHOU INTELLIGENT TECH CO LTD
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Aiming at the defects in the prior art that when the disk in the above-mentioned distributed cluster is abnormal, the reading and writing performance decreases, resulting in too slow response and affecting the user experience, the present invention provides a distributed cluster performance optimization method, device, terminal and storage medium , to solve the above technical problems

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
  • Distributed cluster performance optimization method and device, terminal and storage medium
  • Distributed cluster performance optimization method and device, terminal and storage medium
  • Distributed cluster performance optimization method and device, terminal and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] Such as figure 1 As shown, the present invention provides a method for optimizing distributed cluster performance, including the following steps:

[0054] S 1. Regularly count and record the read and write performance of each disk in the distributed cluster;

[0055] S2. Regularly mark the disks whose read and write performance exceeds the performance threshold in the distributed cluster as slow disks;

[0056] S3. Periodically remove the disks whose number of slow disk marks exceeds the mark threshold from the distributed cluster.

Embodiment 2

[0058] Such as figure 2 As shown, the present invention provides a method for optimizing distributed cluster performance, including the following steps:

[0059] S 1. Regularly count and record the read and write performance of each disk in the distributed cluster; the specific steps are as follows:

[0060] S 11. Regularly count the read-write delay and read-write bandwidth of each disk in the distributed cluster;

[0061] S 12. Store the read-write delay and read-write bandwidth of each disk in the performance database;

[0062] S2. Regularly mark the disks whose read and write performance exceeds the performance threshold in the distributed cluster as slow disks; the specific steps are as follows:

[0063] S21. Read and write time delay and read and write bandwidth of each disk in the performance database;

[0064] S22. Obtain a disk whose read-write delay exceeds the delay threshold, which is defined as a high-latency disk;

[0065] S23. Determine whether the read and...

Embodiment 3

[0073] Such as image 3 As shown, the present invention provides a distributed cluster performance optimization device, including:

[0074] The disk management module 1 is used to regularly count and record the read and write performance of each disk in the distributed cluster; the disk management module 1 includes:

[0075] The disk performance statistics unit 1.1 is used to regularly count the read-write delay and read-write bandwidth of each disk in the distributed cluster;

[0076] The disk performance storage unit 1.2 is used to store the statistical read-write delay and read-write bandwidth of each disk into the performance database;

[0077] The slow disk detection module 2 is used to regularly mark the disks whose read and write performance exceeds the performance threshold in the distributed cluster as slow disks; the slow disk detection module 2 includes:

[0078] The performance reading unit 2.1 is used to read the read-write delay and read-write bandwidth of each...

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 provides a distributed cluster performance optimization method and device, a terminal and a storage medium. The method comprises the following steps: S1, performing statistics and recording on read-write performance of each disk in a distributed cluster at regular time; S2, performing slow disk marking on the disks of which the read-write performance exceeds a performance threshold in a distributed cluster at regular time; and S3, regularly moving the disks of which the slow disk marking times exceed the marking threshold out of the distributed cluster. By monitoring the read-write response time delay of each disk through storage, the slow disks with high long-time read-write time delay and low read-write performance are kicked out of the cluster regularly, so that the influence of the slow disks on the read-write performance of the cluster is reduced, the overall read-write performance of the cluster is improved, and the problem that the overall performance of the cluster is reduced due to low performance of individual disks is prevented.

Description

technical field [0001] The invention belongs to the technical field of cluster performance optimization, and in particular relates to a distributed cluster performance optimization method, device, terminal and storage medium. Background technique [0002] The performance of distributed clusters depends on the performance of each mainstream storage device, and the read and write performance of mainstream storage devices is related to the performance of the underlying disk. When some disks are abnormal, the read and write performance of the abnormal disks will drop, causing the user to respond too slowly when calling to read data, affecting the user experience. [0003] This is a deficiency of the prior art. Therefore, it is very necessary to provide a distributed cluster performance optimization method, device, terminal and storage medium for the above-mentioned defects in the prior art. Contents of the invention [0004] Aiming at the defects in the prior art that when th...

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 Applications(China)
IPC IPC(8): G06F3/06
CPCG06F3/0611G06F3/0613G06F3/0629G06F3/067G06F3/0689
Inventor 陈莹昊李昂
Owner INSPUR SUZHOU INTELLIGENT TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products