Load Balancing in Distributed Systems with Large Time Delays: Theory and Experiment

In distributed-computing environments with high communication delays and cost, limiting the number of balancing instants in scheduling algorithms results in an improved performance compared to the case where load balancing is executed continuously. Therefore, finding the optimal number of balancing instants and optimizing the performance over the inter-balancing time and loadbalancing gain becomes an important problem. In this paper, the performance of a previously reported single load-balancing strategy on a distributed physical system is studied. The performance is also compared to simulations generated by an in-house wireless distributed system. Based on the concept of regeneration in stochastic processes, a mathematical model for the analysis of the queuing systems representing a distributed system with two nodes is developed.