ACO based deployment optimization for software in clouds

Intelligent computing has been employed in deployment optimization for software in clouds to get rid of requirements on rich deployment optimization experience and a number of attempts. However, valuable traditional experience on deployment optimization is not considered to improve the efficiency of searching for deployment options. In this paper, an ACO based deployment optimization approach for software in clouds employing traditional experience is proposed to overcome the deficiency. Targeting cloud environment at production level, the deployment optimization for software in clouds is modeled as a multi-objective ant colony optimization problem; the experience is utilized as heuristics to help behavior decision of ants for more efficient searching; moreover, specific behavioral constraints for ants are designed to drive the algorithm. Finally, experiments comparing with ACO algorithm without experience utilization are conducted, and the results show that our approach outperforms the compared algorithms.