求解云计算任务调度的粒子群优化算法研究
Research on Particle Swarm Optimization Algorithm for Solving Cloud Computing Task Scheduling
DOI: 10.12677/CSA.2018.83033, PDF,  被引量    国家自然科学基金支持
作者: 王 晴*, 付学良, 董改芳, 赵莎莎：内蒙古农业大学计算机与信息工程学院，内蒙古 呼和浩特
关键词: 任务调度；粒子群算法；相关性；执行时间；代价消耗；Task Scheduling； PSO Algorithm； Correlation； Execution Time； Cost Consumption
摘要: 目前，云环境下任务调度问题是一个研究热点，而粒子群算法(Particle Swarm Optimization, PSO)是解决任务调度问题的重要智能算法。针对相关性粒子群算法(Correlation Particle Swarm Optimi-zation, CPSO)和新的基于自适应惯性权重粒子群算法(New Adaptive Inertia Weight Based Particle Swarm Optimization, NewPSO)在解决该问题时易陷入局部最优解和寻优能力差的不足，本文以任务执行时间和代价为目标，将随机因子与惯性权重相融合，提出增强型粒子群算法(Enhanced Particle Swarm Optimization, EPSO)。仿真结果表明，在相同条件下，与PSO算法、CPSO算法和NewPSO算法相比较，EPSO算法能更有效的减少执行时间，降低代价消耗(包括任务执行时间，时间花费以及虚拟机花费)，得到更优的调度方案。
Abstract: At present, task scheduling problem in cloud environment is a hot research topic, and particle swarm optimization algorithm (PSO) is an important intelligent algorithm to solve the task sched-uling problem. According to the correlation, particle swarm Optimization algorithm (CPSO) and new adaptive inertia weight based particle swarm optimization algorithm (NewPSO) in solving this problem are easy to fall into the local optimal solution and poor searching ability, In this paper, the task execution time and cost as the goal, the random factor and the inertial weight are fused, and the Enhanced Particle Swarm Optimization (EPSO) is proposed. Simulation results show that under the same conditions, compared with PSO algorithm, CPSO algorithm and NewPSO algorithm, EPSO algorithm can reduce execution time and cost more effectively (including task execution time, time cost and virtual machine cost), and get a better scheduling solution.
文章引用：王晴, 付学良, 董改芳, 赵莎莎. 求解云计算任务调度的粒子群优化算法研究[J]. 计算机科学与应用, 2018, 8(3): 286-295. https://doi.org/10.12677/CSA.2018.83033

参考文献

[1] Abadi, D.J. (2010) Data Management in the Cloud: Limitations and Opportunities. IEEE Data Engineering Bulletin, 32, 3-12.
[2] Lee, Y.C. and Zomaya, A.Y. (2012) Energy Efficient Utilization of Resources in Cloud Computing Systems. Journal of Supercomputing, 60, 268-280. [Google Scholar] [CrossRef] 
[3] Kennedy, J., Eberhart, R.C. and Shi, Y. (2001) Swarm Intelligence. Elsevier Science Press, Singapore, 202-210.
[4] 何明慧, 徐怡, 王冉, 胡善忠. 改进的粒子群算法优化神经网络及应用[J/OB]. 计算机工程与应用, 2018, 1-9.
http://kns.cnki.net/kcms/detail/11.2127.TP.20180205.1504.006.html
[5] 韩红桂, 卢薇, 乔俊飞. 一种基于种群多样性的粒子群优化算法设计及应用[J]. 信息与控制, 2017, 46(6): 677-684.
[6] 李学俊, 徐佳, 朱二周, 等. 任务调度算法中新的自适应惯性权重计算方法[J]. 计算机研究与发展, 2016, 53(9): 1990-1999.
[7] 滕志军, 吕金玲, 郭力文, 王志新, 许恒, 袁丽红. 基于动态加速因子的粒子群优化算法研究[J]. 微电子学与计算机, 2017, 34(12): 125-129.
[8] 谭文安, 查安民, 陈森博. 优化粒子群的云计算任务调度算法[J]. 计算机技术与发展, 2016, 7(26): 6-10.
[9] Arfeen, M.A., Pawlikowski, K. and Willig, A. (2011) A Framework for Resource Allocation Strategies in Cloud Computing Environment. Proceedings of 2011 IEEE 35th Annual Computer Software and Applications Conference Workshops, 18-22 July 2011, Washington, DC, 261-266.
[10] Netjinda, N., Sirinaovakul, B. and Achalakul, T. (2014) Cost Optional Scheduling in IaaS for Dependent Workload with Particle Swarm Optimization. The Journal of Supercomputing, 68, 1579-1603. [Google Scholar] [CrossRef] 
[11] Embrechts, P., Mcneil, A.J. and Straumann, D. (2001) Correlation and Depend-ency in Risk Management: Properties and Pitfalls. Proc. of the Risk Management: Value at Risk and Beyond, Cambridge University Press, Cambridge, 176-223.
[12] Nelsen, R.B. (2006) An Introduction to Copulas. 2nd Edition, Springer-Verlag, New York, 10-28.
[13] Fard, H., Prodan, R. and Fahringer, T. (2013) A Truthful Dynamic Workflow Scheduling Mechanism for Commercial Multicloud Environments. IEEE Trans on Parallel and Distributed Systems, 24, 1203-1212. [Google Scholar] [CrossRef]