Abdulhamid Shafi'i Muhammad, Abd Latiff Muhammad Shafie, Abdul-Salaam Gaddafi, Hussain Madni Syed Hamid
Faculty of Computing, Universiti Teknologi Malaysia, Johor Bahru, Malaysia.
Department of Cyber Security Science, Federal University of Technology, Minna, Nigeria.
PLoS One. 2016 Jul 6;11(7):e0158102. doi: 10.1371/journal.pone.0158102. eCollection 2016.
Cloud computing system is a huge cluster of interconnected servers residing in a datacenter and dynamically provisioned to clients on-demand via a front-end interface. Scientific applications scheduling in the cloud computing environment is identified as NP-hard problem due to the dynamic nature of heterogeneous resources. Recently, a number of metaheuristics optimization schemes have been applied to address the challenges of applications scheduling in the cloud system, without much emphasis on the issue of secure global scheduling. In this paper, scientific applications scheduling techniques using the Global League Championship Algorithm (GBLCA) optimization technique is first presented for global task scheduling in the cloud environment. The experiment is carried out using CloudSim simulator. The experimental results show that, the proposed GBLCA technique produced remarkable performance improvement rate on the makespan that ranges between 14.44% to 46.41%. It also shows significant reduction in the time taken to securely schedule applications as parametrically measured in terms of the response time. In view of the experimental results, the proposed technique provides better-quality scheduling solution that is suitable for scientific applications task execution in the Cloud Computing environment than the MinMin, MaxMin, Genetic Algorithm (GA) and Ant Colony Optimization (ACO) scheduling techniques.
云计算系统是位于数据中心的大量相互连接的服务器集群,通过前端接口根据客户需求动态分配资源。由于异构资源的动态特性,云计算环境中的科学应用调度被认为是NP难问题。最近,许多元启发式优化方案已被应用于解决云系统中应用调度的挑战,但对安全全局调度问题的关注较少。本文首先提出了使用全球联赛冠军算法(GBLCA)优化技术的科学应用调度技术,用于云环境中的全局任务调度。实验使用CloudSim模拟器进行。实验结果表明,所提出的GBLCA技术在完工时间上产生了显著的性能提升率,范围在14.44%至46.41%之间。从响应时间的参数测量来看,它还显著减少了安全调度应用程序所需的时间。鉴于实验结果,与MinMin、MaxMin、遗传算法(GA)和蚁群优化(ACO)调度技术相比,所提出的技术为云计算环境中的科学应用任务执行提供了质量更好的调度解决方案。