PRESTO: A Parallel Runtime Environment for Scalable Task-Oriented Computations
Datum
2013-10-03
Autor:innen
Zeitschriftentitel
ISSN der Zeitschrift
Bandtitel
Verlag
University of Oregon
Zusammenfassung
While the message-passing paradigm, seen in programming models such as MPI and UPC, has provided a solution for efficiently programming on distributed memory computer systems, this approach is not a panacea for the needs of all scientists. The traditional method of developing parallel applications in C/C++ and Fortran potentially leaves behind high-level and heterogeneous environments which are the most conducive for supporting
modern computational science endeavors. PRESTO alleviates this problem with an easy-to-use framework which provides multi-language adapters to a flexible MPI middleware supporting common computational models such as the asynchronous master/worker and ring pipeline in heterogeneous environments.