In cloud control systems,generating an efficient and economical workflow scheduling strategy for deadline-constrained workflow applications,especially in uncertain multi-workflow dynamic scheduling processes,is a cruc...In cloud control systems,generating an efficient and economical workflow scheduling strategy for deadline-constrained workflow applications,especially in uncertain multi-workflow dynamic scheduling processes,is a crucial challenge.To optimize the total cost of workflow scheduling,the authors propose a cost-driven heuristic scheduling algorithm F-MWSA which consists of two phases:Fuzzy deadline distribution and fuzzy task scheduling.In the fuzzy deadline distribution phase,a new workflow deadline distribution strategy with fuzziness is designed to obtain the sub-deadline constraint of each task.The fuzzy task scheduling phase focuses on a cost-effective strategy to assign tasks to cloud resources,reducing multi-workflow scheduling costs.Performance evaluations on five real-world workflows demonstrate that the proposed F-MWSA outperforms the baseline policy in terms of total cost,success ratio,resource utilization,and makespan.展开更多
基金supported by the National Natural Science Foundation of China under Grant No.62303066the Fundamental Research Funds for the Central Universities under Grant No.2023RC46.
文摘In cloud control systems,generating an efficient and economical workflow scheduling strategy for deadline-constrained workflow applications,especially in uncertain multi-workflow dynamic scheduling processes,is a crucial challenge.To optimize the total cost of workflow scheduling,the authors propose a cost-driven heuristic scheduling algorithm F-MWSA which consists of two phases:Fuzzy deadline distribution and fuzzy task scheduling.In the fuzzy deadline distribution phase,a new workflow deadline distribution strategy with fuzziness is designed to obtain the sub-deadline constraint of each task.The fuzzy task scheduling phase focuses on a cost-effective strategy to assign tasks to cloud resources,reducing multi-workflow scheduling costs.Performance evaluations on five real-world workflows demonstrate that the proposed F-MWSA outperforms the baseline policy in terms of total cost,success ratio,resource utilization,and makespan.
