Design, Performance

Scientific applications are often complex collections of many large-scale tasks. Mature tools exist for describing task-parallel workflows consisting of serial tasks, and a variety of tools exist for programming a single data-parallel operation. However, few tools cover the intersection of these two mod-els. In this work, we extend the load balancing library ADLB to support parallel tasks. We demonstrate how applications can easily be composed of parallel tasks using Swift dataflow scripts, which are compiled to ADLB programs with perfor-mance comparable to hand-coded equivalents. By combin-ing this framework with data-parallel analysis libraries, we are able to dynamically execute many instances of a parallel data analysis application in support of a parameter explo-ration workload