Campus Network Load Balancing Scheduling Algorithm

With rapidly prevalence of Internet, the number of users and Internet-flow also greatlyincreases, which cause the overload operation of the Server. Load balancing scheduler isan effective way of solving the problem. Load balancing uses many servers to provideservices in the network, assigns systematic load to different servers according to certainalgorithm mechanism, and can offer service of simultaneous access for a large number ofusers. The method can also offer stronger performance of scalability and throughput.Schemes of Load balancing scheduler list as follows: RR-DNS, Proxy Server, SolutionBased on Application or IP layer, and so on. How to adopt rational scheduling algorithm tobalance load of some servers is key content of researches of presentNetwork-Load-Balances-Problems and the paper. At present, the common load balancingscheduling algorithm includes: RRS, WRRS, LCS, WLCS, DAHS, SAHS, etc.The main research contents in paper are showed as follows:(1) The paper has studied the structures of Server load balancing, and analyzed factorsof influence on load balancing algorithms, meanwhile exhaustive presented theimplementation techniques of Server load balancing and methods for balancing load basedon hardware and software. This has important directive significance on study of loadbalancing scheduling algorithm.(2) This paper also has researched configurations with the function modules whichmonitor the state parameter of the Server. Implementation methods are analyzed in thepaper, which utilize monitoring system to monitor the flow of the server network, CPUload, and rate of using memory. The research offers the state parameters for load balancingscheduler.(3) Characteristics of Server load and common load balancing scheduling algorithmsare analyzed in paper. Self-adaptation load scheduling algorithm is proposed, and Serverschedule resources according to the states such as Server CPU load, flow of network,memory occupancy etc. Compared with the common load balancing scheduling algorithmswe analyzed and appraised the performances of algorithm presented. The algorithm mayeffectively improve the throughput and shorten response time in heavier loads. Finally, thepaper has discussed the question that need to be probed into and further investigated in thefuture.Utilizing load balancing technology to balance Servers-Load is an effective methodthat solves problems of servers overload operation in the network system structure atpresent. It can increase bandwidth and throughput of Server, and enhance the responsespeed of the Server. It also can shorten the response time, and improve the response qualityof the Server.