A Fuzzy Scheduling Strategy for Online Multi-workflows in IaaS Clouds

With the development of cloud computing, more and more complex workflows have been migrated to clouds and executed with cloud resources. In clouds, it is a a challenging issue to minimize workflow execution cost for scheduling workflow with the deadline constraint. Existing studies are mainly done subject to rigorous conditions without fluctuations, ignoring the fact that workflow scheduling is typically present in uncertain environment. In this paper, we propose a workflow fuzzy scheduling algorithm (WFSA) which aims to minimize workflow scheduling cost and satisfy deadline constraints. In WFSA, we adopt the triangular fuzzy numbers (TFNs) to represent the task processing time and rank tasks to form a scheduling order by the fuzzy sorting strategy. A fuzzy deadline assignment method based on partial critical path (PCPs) of workflows is designed to decomposes the workflow deadline constraint into tasks sub-deadline constraints. A cost-driven fuzzy scheduling strategy is proposed to heuristically select virtual machines (VMs) for minimizing workflow scheduling cost and meeting the task sub-deadline constraint. The experimental results show that our algorithm can effectively reduce the workflow execution cost, compared with other benchmark solutions.