Scheduling Deadline Constrained Task in Hybrid IaaS Cloud using Cuckoo Driven Particle Swarm Optimization

Abstract

Recent advancement in public cloud allow the providers to rapidly diversify their products and pricing plans. To increase cloud provider’s profit ,tasks are outsourced from internal cloud to External Cloud (EC). The main objective is to maximize the internal cloud utilization and minimize the cost of outsourced task to external cloud. This problem is formulated as an integer programming model and is solved using cuckoo driven Particle Swarm Optimization (CS-PSO) approach. This perform the local search more efficiently and it avoids the local optima problem of PSO. The task utilization runtime and the cost for each task is calculated using cloudsim simulator. Experimental results shows that the average profit obtained by this approach is higher than the standard PSO and Self adaptive Learning PSO (SLPSO) for problems of non trivial size.