Data Center Resource Provisioning Using Particle Swarm Optimization and Cuckoo Search: A Performance Comparison

Abstract

The major concerns of cloud providers is resource management. A wide range of approaches and tools have been proposed such as meta-heuristics. Among the most recent suggested meta-heuristics are: Cuckoo Search (CS) and Particle Swarm Optimization (PSO). The main contribution of this paper is to compare the performance of CS and PSO on the resource allocation in a data center. Extensive simulation, using a dataset varying from the range of 200 to 1000 demands, demonstrates that PSO converges most rapidly than CS. Moreover, the results show an enhancement of as much as 1% to 10% of energy consumption and 1% to 5% of resource utilization on average.