Task Allocation in Workflow Scheduling on Parallel Computing Platform

• Main Authors: Tsai, Ying-Lin, 蔡英麟
• Other Authors: Kuo-Chan Huang
• Format: Others
• Language: en_US
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/00582246861932675064

碩士 === 國立臺中教育大學 === 資訊工程學系 === 101 === With the advancement of technology and emergence of grid and cloud computing, now many large-scale scientific and engineering applications are usually constructed as workflows due to large amounts of interrelated computation and communication. Many approaches have been proposed to deal with the challenging workflow scheduling problem. List scheduling and clustering are the two most common types of workflow scheduling heuristics. In this thesis, we make contributions to these two types of workflow scheduling, respectively. In the first part, we developed new task ranking and allocation methods for list-based scheduling approaches. In the second part, we proposed efficient task group allocation methods, considering both resource fitness and tasks’ EFT (Earliest Finish Time), for clustering-based concurrent workflow scheduling. The proposed approaches were evaluated through a series of simulation experiments and compared to typical workflow scheduling methods. The experimental results show that our approaches outperform existing methods significantly, achieving up to11.8 % and 15.5% performance improvement in terms of average makespan for list-based and clustering-based approaches, respectively.