Method and system for realizing server load balance

Abstract

The invention discloses a method and system for realizing server load balance. The method for realizing server load balance comprises the steps of: at a client side, sending a request for calling actual servers; at an actual server side, acquiring state information of the actual servers according to configuration, calculating the score of the load of the current actual server according to the state information, and reporting a corresponding scoring result; and, at a distribution server side, receiving scoring results of loads of all the actual servers, furthermore, dividing the priority processing levels of all the actual servers according to the scoring results, and, when the request for calling the actual servers and sent by the client side is received, distributing the actual servers corresponding to the priority processing levels according to the client permission. By means of the technical scheme in the invention, each actual server can be sufficiently used; furthermore, overloadof the actual server, which is poor in performance, can be avoided; a client having high priority can be rapidly responded; and thus, the user experience is improved.

Description

technical field [0001] The invention relates to a communication technology method, in particular to a method and system for realizing server load uniformity. Background technique [0002] The current general implementation of load balancing is: load balancing is a set of servers composed of multiple servers in a symmetrical manner, each server has an equivalent status, and can provide services independently without the assistance of other servers. Through a certain load sharing technology, the requests sent from the outside are evenly distributed to a certain server in the symmetrical structure, and the server that receives the request independently responds to the client's request. Load balancing can evenly distribute customer requests to the server array, thereby providing fast access to important data and solving the problem of a large number of concurrent access services. [0003] Currently common load balancing algorithms include DNS-based load balancing, load balancin...

AI Technical Summary

Problems solved by technology

Its advantages are simple implementation, easy implementation, low cost, and is suitable for most TCP / IP applications; however, its disadvantages are also very obvious. First of all, this solution is not a real load balancing, and the DNS server evenly distributes Http requests to the background Web server, regardless of the current load of each Web server; if the configuration and processing capacity of the background Web servers are different, the slowest Web server will become the bottleneck of the system, and the server with strong processing capacity cannot give full play Function; secondly, fault tolerance is not considered. If a web server in the background fails, the DNS server will still distribute DNS requests to the failed server, resulting in failure to respond to the client

The last point is fatal, it may cause a considerable number of customers not to enjoy web services, and due to DNS caching, the consequences will last for a long time (generally, the refresh cycle of DNS is about 24 hours)

Images