An Improved Estimation of Distribution Algorithm for Cloud Computing Resource Scheduling

This paper focuses on cloud computing resource scheduling on the Soft as a Service layer and aims at minimizing the user costs by regarding the deadline as a constraint for scheduling independent tasks. Existing works with evolutionary computation approaches fail to describe the interactions among independent tasks. To overcome this problem, an improved Markov-chain-based estimation of distribution algorithm is proposed, and the concept of virtual machine selection diversity is created to construct the probabilistic model rationally. Moreover, one heuristic rule related to the investigated problem is created to keep the population maintaining a high diversity in the evolution process. The experiment results show that the proposed algorithm not only obtains the best solution quality but also has competitive convergence among all compared algorithms.