Abstract:
Cloud brokers can help consumers to discover and select suitable cloud computing services from a variety of different cloud providers. With the increase in the scale and complexity of application systems, it is becoming a challenge for cloud brokers to select the optimal cloud services and deploy them in multiple cloud providers to effectively mitigate the problem of vendor lock-in. Genetic algorithm and simulated annealing algorithm are two possible candidates for the optimization problem. Although genetic algorithm has a strong global search ability, it easily falls into the local optimal solution. Simulated annealing algorithm has strong local search ability and can jump out of the local optimal solution, but its efficient is lower. Aiming at the above shortcoming, this paper proposes a simulated annealing genetic algorithm by utilizing the powerful global search capability of genetic algorithm in the early stage. At the latter stage, this proposed algorithm uses simulated annealing algorithm to obtain the global optimal solution. Moreover, this paper presents a cloud resource scheduling method based on simulated annealing genetic algorithm for searching the resources that meet the demands of QoS applications in the cloud broker. Experiment results show that the proposed approach can effectively increase the convergence speed and improve the efficiency of the algorithm without affecting the solution precision.