An Optimal Engine Component Placement Strategy for Cloud Workflow Service

Workflows have been used to represent a variety of applications that involve coordinating a set of business services or scientific services, which are generally geographically distributed. With the development of cloud computing, a workflow engine can be deployed as a cloud service, responsible for executing customers' workflow instances. In a cloud workflow service, workflow engine components can be placed into different cloud regions. Thus, one challenging problem that arises is how to select the appropriate cloud regions to place the workflow engine components in order to efficiently execute a service workflow instance. Because this is a typical nondeterministic polynomial-time hard (NP-hard) problem, we propose a heuristic algorithm to select the regions where to place workflow engine components in an optimal and efficient way, with the objective of reducing the execution time of the service workflow instance. The experimental results prove that our proposed algorithm has higher performance than other approaches in terms of the solution quality and the running speed.