Catfish-PSO based scheduling of scientific workflows in IaaS cloud

详细信息    查看全文

• 作者：S. Jaya Nirmala ; S. Mary Saira Bhanu
• 关键词：Cloud computing ; Job scheduling ; Catfish particle swarm optimisation ; Scientific workflows ; WorkFlowSim
• 刊名：Computing
• 出版年：2016
• 出版时间：November 2016
• 年：2016
• 卷：98
• 期：11
• 页码：1091-1109
• 全文大小：1,724 KB
• 刊物类别：Mathematics and Statistics
• 刊物主题：Mathematics
  Computational Mathematics and Numerical Analysis
  
• 出版者：Springer Wien
• ISSN：1436-5057
• 卷排序：98

文摘

Cloud computing is a technology wherein a network of remote servers is used to process large amount of data in real-time. The servers and data sources may be located in geographically distant regions. Scheduling of workflows is one of the major challenging issues in cloud computing. Workflows are used to express a wide variety of applications including scientific computing and multi-tier web applications. The Workflow scheduling problem is known to be NP-complete. No known traditional scheduling algorithm is able to provide an optimal solution in polynomial time for NP-complete problems. So, researchers rely on heuristics and meta-heuristics to achieve the most efficient solution. In this paper, a workflow scheduling algorithm is proposed to schedule large scientific workflows that are to be executed on IaaS clouds. The workflow scheduling algorithm generates a schedule with the task-to-resource mapping. The metaheuristic Catfish particle swarm optimization (C-PSO) technique is used to select the best schedule with the least makespan and execution cost. The performance of C-PSO is then compared with traditional PSO. The algorithm is simulated on the WorkFlowSim Simulator, an extension of CloudSim simulator. The solution is tested for different types of scientific workflows like Montage, Epigenome, CyberShake and Inspiral. It is observed from the experimental results that C-PSO gives better performance than traditional PSO in terms of execution cost and makespan.