A Deadline-Constrained and Cost-Minimized approach for Workflow scheduling in IaaS clouds
Keke Chen, Yan Wang, Lei Zhang, Gangzhi Xie · 2022 IEEE 6th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC ) · 2022
Workflow is widely used to represent and model complex applications of various scientific and business fields. Due to the compute-intensive and communication-intensive character-istics, cloud computing, which can provide a great quantity of resources at short notice, is undoubtedly an ideal deployment environment. In cloud computing, cost and time are two major concerns for workflow scheduling problems. Various algorithms have been devised for this problem; however, they are not satisfactory in terms of generating schedules within acceptable time. In view of this, this paper studies the deadline-constrained and cost-minimized workflow scheduling problem, and proposes a well-designed approach named Resource Integration Workflow Scheduling(RIWS). RIWS consists of four phases: workflow partitioning, deadline distribution, task selection and service selection, which accounts for partitioning tasks into dependency free sets, allocating sub-deadlines for each task, establishing the scheduling order of tasks and assigning services for tasks, respectively. The results demonstrate that, RIWS has higher scheduling success rate and can find less expensive solutions for the testing workflows when compared to two peer approaches.