Automated Negotiation using Parallel Particle Swarm Optimization for Cloud Computing Applications

Negotiating service level agreements (SLAB) between cloud consumers and service providers is a key aspect of highly automated cloud systems. However, this process needs to be optimized to reduce the time of negotiation and the quality of solutions. In this paper, a parallel implementation of the particle swarm optimization (PSO) technique is proposed to improve the SLA negotiation in cloud computing by reducing the negotiation time and increasing the throughput. The paper highlights the main steps of designing and implementing two Parallel PSO SLA Negotiation algorithms (Synchronous and Asynchronous algorithms). The algorithms are compared with sequential PSO SLA Negotiation to evaluate their effectiveness and performance. Our results satisfy a speedup of up to 30% and an increase in throughput of 15% The Asynchronous algorithm yields better results than the Synchronous one (30% quicker negotiation time and 10% in the throughput).