Adaptive Load Balancing Algorithm Using Service Queue
G. Kanagaraj, Naveen Shanmugasundaram, Sathish Prakash · 2012
Abstract— In this internet driven age it is difficult to meet the demands of the people in E commerce. This is fixed by using the load balancing techniques. Load balancing refers to distributing incoming HTTP requests across web servers in a server farm. This is achieved by various load balancing algorithms like Round robin, Weighted Round robin, Observed, Fastest etc. Even though objective of load balancing is achieved in recent years; still placing of computational logic and decision making and zeroing in for the best server has not attained its efficiency. It is under study. We suggest a method to dynamically load balance using service queue wherein every server computes its load value by summing the load parameters like memory utilization, CPU utilization, and network utilization and exchange load value with central node in a certain cyclic period. Central node selects the least loaded server among farm of servers to process the request. Each server at central node waits in a queue called service queue for its turn to process the request.