DESIGNING A LOAD BALANCING FRAMEWORK FOR EFFICIENT SERVER-SIDE COMPUTATION ACCESS

Abstract

The computational (server) environment has been faced with several challenges during operations of the cooperating computing systems with adverse outcome. The resultant effect is the gradual slowdown of the performance of the system in attending to user requests, drop in response time, hang and crash issues among others. Some of these issues occur when the server is no longer responding to users request or the server becomes irresponsive to user actions; all these cause the system to fail. Deadlock during computation is also an aspect of the challenges caused on the server as different users compete for resources held by other users. This research presents a framework for implementation of load balance access in a computational (server) environment. This is achieved by identifying factors leading to load imbalance, determining the degree of imbalance, developing a framework that will reduce imbalance. Computing an average neighborhood load balancing using a 10% trigger on the server while checking for situations when the server is overloaded, under loaded and loaded. The average neighborhood was used to achieve this and the system was implemented with visual C#. The system achieved an efficient load balancing process using the average neighborhood load balancing method.