Expandable distributed coordination service management method under cloud platform

Abstract

The present invention discloses an expandable distributed coordination service management method under a cloud platform, which comprises the steps of: initializing all nodes of a distributed cluster in an organization manner of middle node crossing forest; inquiring father member node and son member node information of each node of all member nodes according to a global network topology structure from a central management node; performing information synchronization and communication among all father member nodes and son member nodes according to a heartbeat mechanism; determining whether a member state variation information is received from other member nodes by each member node; if the member state variation information is received, further determining whether the member state variation information indicates number reduction or increase of members; and if number reduction of members is determined, after the member node receives member state variation information which indicates number reduction of members, packaging a piece of member state variation information according to heartbeat interruption, and furthermore transmitting the member state variation information to the central management node. According to the expandable distributed coordination service management method, service updating efficiency does not reduce rapidly in the expanding process, and therefore the expandable distributed coordination service management method is more suitable for a large-scale cloud platform and data processing system.

Description

technical field [0001] The invention belongs to the technical field of cloud computing, and more specifically relates to a scalable distributed coordination service management method under a cloud platform. Background technique [0002] Today, when cloud computing and big data are prevalent, cloud services and data processing and analysis are usually built on machine platforms with a scale of tens of thousands or even millions. In a cluster of such a scale, problems such as node failure, disk error, and even network failure have a high probability of occurrence, which poses a great challenge to service fault tolerance and reliability. Coordination services are usually used to solve the problem. The coordination service system is the best solution to solve the problem of fault tolerance and ensure reliable service without changing the original system architecture. Through the reliable coordination service, it can realize the reliable management of clusters, distributed algori...

AI Technical Summary

Problems solved by technology

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides a scalable distributed coordination service management method under the cloud platform, the purpose of which is to solve the poor scalability and low efficiency of the management process existing in the existing coordination service system technical issues

Images