Research On Load Balancing Algorithm And Its Application In Electronic Commerce Platform

With the rapid development of Internet technology, more and more users use the network service, which means that the increase of network traffic, a variety of con-current degree are on the rise, these factors are put forward higher requirements on the network service provider, for some high load Web service server, processing ca-pacity CPU and I/O gradually cannot meet user’s requirements. From the develop-ment of network technology, increase memory access speed growth, server processing speed can not catch up with the bandwidth of the growth rate, which means that the e-commerce platform, will be a performance bottleneck more appear. There are two ways to solve this problem, upgrading hardware and build server cluster load balanc-ing. The construction method of server cluster load balancing is better than the first method, in both cost and performance optimization of so, this paper mainly studies the server performance and building a cluster of related technology of load balancing.Optimization study and to some extent Web server Apache2.0for this paper, but also on the Linux virtual server (LVS) and load balancing algorithms are studied and analyzed, and puts forward a load balancing algorithm based on dynamic feedback.In this paper, based on the analysis of the Apache server, using some methods to optimize the performance of Apache server in practice. To achieve load balancing cluster of Web servers using the Linux virtual server LVS; this paper proposes a dy-namic load balancing algorithm based on feedback, and made a simulation experi-ment on the. To alleviate the increasing load of WEB server, the load balancing is one of the most effective methods, implementation of this method to the maximum extent possible to use existing resources to improve server performance, improve resource utilization. On the existing network structure, the server cluster load balancing im-plementation can effectively improve the usability of the system, to avoid a single point of failure, to provide better protection for the stable operation of the electronic commerce platform, to the user experience better application.At present, in this paper, the Apache optimization and LVS cluster to do detailed analysis based load balancing technology, make the simulation experiment in an auto parts electronic commerce platform, to optimize the performance of the proposed and load balancing algorithm based on dynamic feedback after the test and data analysis and contrast, and satisfactory results.