Provided by: Science and Development Network (SciDev.Net)
Date Added: Apr 2012
The cluster web server does not only allow clients to access a site that will provide better response time, but also provides an another level of redundancy in the case of the whole site failure. In advance of a simple load balancing, this requires a comprehensive scheme for traffic management, which includes the principle of an optimal load-balancing for the process of client requests across multiple clusters of the real servers. This paper presents an optimal load balancing algorithm base on ANFIS (Adaptive Neuro-Fuzzy Inference System) modeling and FIS (Fuzzy Inference Engine) for the local status of the real servers using the performance metrics, and show the benefits provided by the suggested method in the load-balancing performance.