Layering-based distributed cloud computing centre load balancing method

Abstract

The invention relates to a layering-based distributed cloud computing centre load balancing method, which is a load balancing method for a cloud computing centre based on brand new design architecture. With the method, single point of failure is avoided, scale limitation is overcome and overall overloading or underloading of local areas is solved; the method is suitable for a large-scale cloud data centre and can avoid single point of failure, also can solve the problem of overall overloading or underloading of local areas, greatly improves utilization rate of node resources and avoids the problem of single point of failure.

Description

technical field [0001] The invention relates to a layer-based distributed cloud computing center load balancing method, which belongs to the field of cloud computing and cloud load balancing. Background technique [0002] With the development of Internet network technology and the continuous improvement of computer technology, the ability of data transmission and processing in the network has increased linearly. People hope to obtain a direct and convenient way of computing and processing, without installing application software, as long as they are connected to the Internet, they can use the idle computer resources connected to the network to process tasks. [0003] In this context, cloud computing emerged as the times require. The so-called cloud computing is to connect a cloud data center composed of a large number of servers and storage device clusters through a computer network to obtain the services required by remote clients. Multiple virtual machines are set up on t...

AI Technical Summary

Problems solved by technology

But this will lead to a single point of failure problem, and this method cannot adapt to the requirements of large-scale cloud data center load balancing; the improved distributed matching method is based on the distributed matching method and introduces distributed hash tables DHTs and " Adjacent" concept, each node randomly obtains the location and load information of some adjacent nodes, and stores them in DHTs, and each node performs load balancing operations spontaneously in a local area (some nodes) according to the load information in the table

This method can avoid the single point of failure problem and is suitable for large-scale cloud data centers. However, in the case of overall overload or light load in a local area, this method will not be able to effectively load balance

Images