Authors: Ying Zhao; Bin Hu; Zongjie Zhang; Rongkui Zhang
Addresses: Yunnan Electric Power Dispatching Control Center, Kunming, Yunnan 650011, China ' Yunnan Electric Power Dispatching Control Center, Kunming, Yunnan 650011, China ' Yunnan Yundian Tongfang Technology Co., Ltd., Kunming, Yunnan 650017, China ' Yunnan Yundian Tongfang Technology Co., Ltd., Kunming, Yunnan 650017, China
Abstract: In order to solve the problems of high energy consumption and time cost of cloud service workflow task scheduling and poor resource utilisation of cloud platform, a cloud service workflow scheduling algorithm based on priority rules is proposed. Firstly, according to the characteristics of cloud service workflow architecture, the workflow instances to be processed are classified by MCUD algorithm. Secondly, according to the classification results, genetic algorithm is used to generate the scheduling strategy to meet the needs of users. Finally, priority rules, namely cloud task priority heuristic rules, are introduced to determine the workflow task order and update it to reduce the completion time and cost of workflow scheduling. The results show that this method can shorten the execution time, reduce the energy consumption cost, and improve the resource utilisation effect of cloud platform, which verifies the feasibility of this method and the effectiveness of the algorithm.
Keywords: MCUD algorithm; genetic algorithm; priority rules; workflow; objective function.
International Journal of Internet Manufacturing and Services, 2022 Vol.8 No.3, pp.254 - 266
Received: 30 Dec 2021
Accepted: 28 Mar 2022
Published online: 18 Jul 2022 *