Using an Actor Framework for Scientific Computing: Opportunities and Challenges

keywords: Scientific workflows, actor model, workflow patterns, Akka framework
We examine the challenges and advantages of using an actor framework for programming and execution of scientific workflows. The following specific topics are studied: implementing workflow semantics and typical workflow patterns in the actor model, parallel and distributed execution of workflow activities using actors, leveraging event sourcing as a novel approach for workflow state persistence and recovery, and applying supervision as a fault tolerance model for workflows. In order to practically validate our research, we have created Scaflow, an Akka-based programming library and workflow execution engine. We study an example workflow implemented in Scaflow, and present experimental measurements of workflow persistence overhead.
reference: Vol. 35, 2016, No. 4, pp. 870–889