Implementation method for providing load balancing service based on OpenStack architecture

Abstract

The invention discloses an implementation method for providing a load balancing service based on an OpenStack architecture. The method comprises the following steps that a client is responsible for sending a network request message to a server; a router is responsible for converting a destination IP address requested by a user into an IP address of an intranet of the server, forwarding the requestmessage to a load balancer, and the load balancer is used for selecting the appropriate server according to a configured processing algorithm after the load balancer receives the request, and sendingthe request message to the server for processing; and the server is an entity for processing the user request and is formed by a plurality of servers providing the same service. The method has the beneficial effects that the method can support load balancer equipment and software load balancers of different manufacturers and the load balance between virtual machines and between the servers, a container or automatic operation and maintenance can be adopted to realize deployment, the maintenance cost is reduced, the flexibility is increased, a cluster type deployment mode can be realized, and the performance and the stability of the load balancer can be improved.

Description

technical field [0001] The invention relates to the technical field of load scheduling in cloud computing, in particular to an implementation method for providing load balancing services based on the OpenStack architecture. Background technique [0002] With the rapid development and maturity of cloud computing technology, more and more enterprises adopt cloud architecture to deploy their business and applications and build test environments, and migrate the application load of traditional data centers to the cloud, which will undoubtedly increase the cloud architecture. load balancing requirements. In addition, due to the characteristics of large system scale, multiple resource types, and multiple task requests, a problem that large-scale cloud computing service operators need to face is how to schedule virtual machines to improve the utilization of physical machine resources and reduce the occurrence of excessively high or low loads. phenomenon to maximize the overall per...

AI Technical Summary

Problems solved by technology

[0004] Problem: In the age of cloud computing, enterprise applications and businesses are gradually becoming cloud-based, and services are migrated to the cloud. It is difficult to use hardware-implemented load balancers to solve the problem of load distribution among virtual machines on the cloud (hardware load balancing The load balancer is usually used in the data center to achieve the purpose of load balancing between servers). In addition, the performance of the load balancer between multiple manufacturers is quite different, and the compatibility between them is also relatively poor.

[0007] 1) The load balancer implemented based on hardware needs to purchase expensive hardware equipment, and also requires redundant configuration, which makes deployment difficult; in addition, professional operation and maintenance personnel are also required, and the hardware and labor costs are high

[0008] 2) The load balancer implemented based on hardware cannot effectively grasp the status of the server and the application. When the system processing capacity is seriously reduced, the network layer may not be aware of it, and the network transmission will still be carried out, eventually causing the server to crash

Images