Argonne National Laboratory

Language Features for Scalable Distributed-Memory Dataflow Computing

TitleLanguage Features for Scalable Distributed-Memory Dataflow Computing
Publication TypeConference Paper
Year of Publication2014
AuthorsWozniak, JM, Wilde, M, Foster, IT
Conference NameProc. Data-Flow Models for Extreme Scale Computing at PACT, 2014
Date Published08/2014
Conference LocationEdmonton, Alberta, Canada
Other NumbersANL/MCS-P5169-0714
AbstractDataflow languages offer a natural means to express concurrency but are not a natural representation of the architectural features of high-performance, distributed-memory computers. When used as the outermost language in a hierarchical programming model, dataflow is very effective at expressing the overall flow of a computation. In this work, we present strategies and techniques used by the Swift dataflow language to obtain good performance on extremely large computing systems. We also present multiple unique language features that offer practical utility and performance enhancements.