A predictive modified round robin scheduling algorithm for web server clusters

As dynamic contents are changing traditional web environments, there is an increasing demand on high performance web servers, which leads to the use of cluster-based web servers. As a result, utilization of cluster web servers through effective and fair load balancing is a crucial task specifically when it comes to advent of dynamic contents and database-driven applications on the internet such as e-commerce and corporate database. After stating the problem discussed in this paper and some preliminaries, we propose a Predictive Modified Round Robin load-balancing algorithm (PMRR), which applies the mythology of prediction into load-balancing field. We validate its effectiveness of proposed load-balancing algorithm through simulation results. The algorithm significantly decreases both the load range and load variance in contrast to Round Robin (RR) scheduling and Modified Round Robin (MRR) scheduling.