Load balancing method and system applied to distributed application system

Abstract

This invention relates to a load balancing method and system applied to a distributed application system. The method comprises the steps of: receiving a request sent by a client; judging whether the client establishes a connection with a server, when no connection is established, selecting a server with lighter load to establishing a connection with the client; when the connection is established between the client and the server, judging whether the load of server is excessive; when the load of server is excessive, releasing the connection of the client with the server and re-selecting other servers with lighter load to establish a connection; and submitting the request to the connected server for execution. The invention avoids that the server can continuously process the request of the client under the condition of excessive load of the server, and ensures balance distribution of the load among the servers, thereby improving the reliability of the load balancing. .

Description

technical field [0001] The present invention relates to the field of information technology, and more specifically relates to a load balancing method and system applied to a distributed application system. Background technique [0002] At present, with the increase of business volume, the rapid growth of access volume and data flow in each core part of the existing network, its processing capacity and calculation intensity also increase accordingly, making it impossible for a single device to bear it. Therefore, the load balancing mechanism came into being. pregnancy. [0003] Load balancing refers to the balanced distribution of loads to multiple computers, network connectors, processors, hardware devices or other resources in a network environment in order to obtain higher resource utilization, improve system throughput, and reduce response time. time to avoid overloading. [0004] An existing load balancing method is to directly select a server with a lighter load when ...

AI Technical Summary

Problems solved by technology

[0004] An existing load balancing method is to directly select a server with a lighter load when the client has business to be processed, and then establish a connection with the service object on the server, and the service object processes and completes the corresponding service sent by the client. business requests, and there are a large number of concurrent operations of clients in a distributed application system, the server that establishes a connection with the client may handle requests of multiple service types, and the load on the server will be seriously unbalanced during the operation of the system phenomenon, and to complete a business, the client may need to send a series of requests, and the client and the service object in the server exchange information based on a connection for a long time, which will also cause the server to be overloaded. Therefore, using With this existing method, there is still the phenomenon of unbalanced server load in the system

Images