Balanced scheduling of distributed workflow tasks based on clustering

作者：

Highlights：

• A runtime balance clustering algorithm (RBCA) is proposed to solve the problem of runtime imbalance.

• A dependency balance clustering algorithm (DBCA) is proposed to solve the problem of data dependency imbalance.

• A strategy of choosing RBCA or DBCA is proposed according to data transmission amount and task runtime.

• The clustering algorithms are compared with existing algorithms through WorkflowSim with extensive experiments.

摘要

•A runtime balance clustering algorithm (RBCA) is proposed to solve the problem of runtime imbalance.•A dependency balance clustering algorithm (DBCA) is proposed to solve the problem of data dependency imbalance.•A strategy of choosing RBCA or DBCA is proposed according to data transmission amount and task runtime.•The clustering algorithms are compared with existing algorithms through WorkflowSim with extensive experiments.

论文关键词：Workflow,Task clustering,Runtime balance,Dependency balance,Scheduling,Dependency correlation,Distributed computing

论文评审过程：Received 28 January 2020, Revised 30 March 2020, Accepted 16 April 2020, Available online 22 April 2020, Version of Record 8 May 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.105930