Dynamic and Decentralized Load Balancing Approach in Small Scale Distributed Systems

Abstract

Abstract:To improve the throughput of small-scale distributed systems, workloads have to be evenly distributed among all the computing nodes of distributed systems. As any distributed application, can undergo critical performance issues due to workload imbalances on small scale distributed system and heterogeneity of resources. The main goal to prevent, if possible, the condition where some processors are overloaded with set of jobs while other processors are lightly loaded, moderately loaded, heavily loaded and some may be idle processors. This is due to the characteristics of distributed systems and complex nature of the users job. In this paper we have proposed real-time prototype model for small-scale distributed systems. Dynamic and decentralized load balancing algorithm has been implemented to evenly distribution of jobs among all the computing nodes of small-scale distributed systems through which to minimize execution time, response time, communication latency and maximize throughput, CPU utilization. It is beneficial for all those researches who want to make more enhance and efficient load balancing techniques for real-time distributed systems in near future.