Cluster election method and system

A clustered, unavailable technology, applied in the field of big data computing, which can solve problems such as inconsistent algorithms, IP conflicts, and failure to consider high-availability cluster processing solutions.

Inactive Publication Date: 2017-05-31
GOSUNCN TECH GRP
View PDF4 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Today's big data operations have their own cluster processing modes, and the algorithms are not uniform
The company's self-developed system does not even have a cluster, which leads to the failure to work normally after the system crashes, affecting the normal execution of big data processing
Moreover, many existing old systems do not consider the high-availability cluster processing scheme, or only use the simple vrrp protoco

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
  • Cluster election method and system
  • Cluster election method and system

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0041] Example 1

[0042] Such as figure 1 Shown is a flowchart of a specific embodiment of a cluster election method of the present invention. See figure 1 , The specific steps of a cluster election method in this specific embodiment include:

[0043] S1. When the slave server cannot obtain the heartbeat of the master server, the status of all slave servers is changed to candidate servers, and a voting request is sent to the remaining servers to be selected as the master server.

[0044] Usually when the system is started or the master server is offline, the slave server cannot obtain the heartbeat of the master server. At this time, the slave server can be given time to wait for the heartbeat of the master server by presetting a certain time threshold or the number of times the master server cannot obtain the heartbeat. When the heartbeat of the master server cannot be obtained after a certain time threshold or more than a certain number of times (for example, the heartbeat of the...

Example Embodiment

[0051] Example 2

[0052] The present invention also provides a cluster election system to realize the high-availability scheme of the non-high-availability system.

[0053] Such as figure 2 As shown, this specific embodiment is a structural diagram of a cluster election system. See figure 2 In this embodiment, a cluster election system specifically includes a server. The server can determine the master server and the slave server through selection or election. When the master server stops the server, the system of the present invention can be used to flush and elect the master server. Provide services. Specifically, the server includes:

[0054] The status change module 110 is used to change the status to a candidate server when the server cannot obtain the heartbeat of the master server; the original status of the server is the status of the slave server, and the slave server cannot get the heartbeat of the master server when the system is started or the master server is offl...

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 the field of large data computing, and more specifically relates to a cluster election method and system for large data operations. The method comprises the following steps: when a slave server cannot obtain the heartbeat of a master server, changing the state of the slave server into a candidate server, and sending a voting request of sending the slave server as the master server to the rest servers; when the candidate server receives the votes of the server exceeding a preset threshold number, electing the candidate server as the master server, and changing the state of the candidate server into the master server; when the candidate server waits for the voting result, electing the other servers as the master server, and changing the state of the candidate server into the slave server; and when no candidate server is elected as the master server, sending the voting request again by the candidate server until a candidate server becomes the master server. According to the cluster election method and system, all kinds of application clusters are formed by a distributed voting protocol, so that the system without high availability has high availability.

Description

technical field [0001] The present invention relates to the field of big data computing, and more specifically, to a cluster election method and system for big data computing. Background technique [0002] Today's big data operations have their own cluster processing modes, and the algorithms are not uniform. The company's self-developed system does not even have a cluster, which causes it to fail to work normally when the system crashes, affecting the normal execution of big data processing. Moreover, many existing old systems do not consider the high-availability cluster processing solution, or only use the simple vrrp protocol for high-availability processing. When the control node or network node in the virtualization system stops serving due to various reasons, the system cannot control the computer network function of the virtualization system, and can only wait for the service to resume. For the solution using vrrp, generally only two machines are in hot standby mod...

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
IPC IPC(8): H04L12/24
CPCH04L41/30H04L41/0668
Inventor 李文杰范日明林楷填毛亮黄仝宇李旭泉宋一兵汪刚侯玉清刘双广
Owner GOSUNCN TECH GRP
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap