The optimization of virtual resource allocation in cloud computing based on RBPSO. (27th December 2018)

• Record Type: Journal Article
• Authors: Wang, Xiaohui; Gu, Haoran; Yue, YuXian
• Other Names: Wu Yulei guestEditor.; Yan Zheng guestEditor.; Zhao Zhiwei guestEditor.; Al‐Dubai Ahmed guestEditor.; Li Tonglin guestEditor.; Xie Bing guestEditor.; Zhang Boyu guestEditor.
• Is Part Of: Concurrency and computation. Volume 32:Number 16(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2018-12-27
• Subjects: cloud computing -- virtual resource allocation -- resampling binary particle swarm optimization; Parallel processing (Electronic computers) -- Periodicals; Parallel computers -- Periodicals; 004.35
• DOI: 10.1002/cpe.5113
• Languages: English
• ISSNs: 1532-0626

Abstract

Summary: The virtual resource allocation in cloud computing is becoming a critical issue. In order to meet the task requirements of different users, virtual machines need to be placed on physical machines through virtualization technology in the data center. However, in this process, the total load balance, energy consumption, and resource utilization of physical machines should be considered. Therefore, two models are established for two different optimization targets, respectively. The first model is built to minimize the degree of load imbalance. The second model is built to maximize the resource utilization and minimize the energy consumption. To gain better results of virtual machines placement, we propose a new algorithm called resampled binary particle swarm optimization (RBPSO). To enhance the global search ability of BPSO, we add the re‐sampling, mutation and small vibration process to it, named RBPSO, for the purpose of maintaining the diversity of the population, reducing redundant calculation and thereby improving the ability and efficiency of the algorithm. Then, the RBPSO is used to solve the deployment problem of virtual machines in cloud computing. The experiments show that the proposed model is reasonable and RBPSO performs better than BPSO and genetic algorithm (GA).