A fuzzy grouping-based load balancing for distributed object computing systems

In the distributed object computing systems a set of server objects are made available over the network for computations on behalf of remote clients. In a typical distributed system setting the existing load balancing algorithms usually do not consider the global system state. In this paper we propose a new approach for improving the performance of distributed system using fuzzy grouping-based load balancing. It utilizes membership graphs in terms of the amount of CPU time and memory used for inferencing the service priority. Extensive computer simulation reveals that the proposed approach allows consistently higher performance than other approaches in terms of response time and throughput for various number of servers and tasks.