期刊名称：International Journal of Advanced Research In Computer Science and Software Engineering
印刷版ISSN：2277-6451
电子版ISSN：2277-128X
出版年度：2013
卷号：3
期号：6
出版社：S.S. Mishra
摘要：In a distributed network of computing hosts, the performance of the system can depend basically how the work is efficiently divide across the participating nodes. Dynamic load balancing have better performing potential than static load balancing, but they are quite complex. The overheads involved in dynamic load balancing are much more. But one cannot neglect their benefits. The major factors that are strongly considered for the performance are load index, queue-length-based indices. Load balancing is very effective when a large portion of the workload is immobile. All hosts, along with lightly loaded nodes get benefit from load balancing. Similarly, all types of jobs get improvements in their response times, with larger jobs benefiting more. System instability is possible, but can be easily avoided. In a distributed system, the random arrival of number of tasks at each processor is likely to bring about uneven processor loads. There are various approaches to provide the dynamic load balancing
