A lightweight communication interface for parallel programming environments
We present a small, extensible software interface for the communication between different parallel programming models. With only four new commands our PLUS communication interface can be easily integrated into existing parallel codes, allowing tasks to transparently communicate from, e.g., PVM to MPI and PARIX, or any other parallel programming model.
PLUS is one important software module that has been developed within the Metacomputer Online initiative. The core idea of Metacomputer Online is to design small, versatile and extensible interfaces between existing software modules with the goal to build a WAN metacomputer by linking suitable existing software packages.
Our current PLUS implementation supports inter-process communication between PVM, MPI and PARIX. Much effort has been spent on optimizing the communication across internet and intranet links. As a result, our PLUS communication is usually faster than raw stream socket TCP/IP communication supported by the various parallel programming models.
Unable to display preview. Download preview PDF.
- 1.Carl Albing. Cray NQS: production batch for a distributed computing world. In Proceedings of the 11th Sun User Group Conference and Exhibition, pages 302–309, Brookline, MA, USA, December 1993. Sun User Group, Inc.Google Scholar
- 2.John Bloomer. Power Programming with RPC. O'Reilly & Associates, 1992.Google Scholar
- 3.T. DeFanti, I. Foster, M. Papka, R. Stevens, and T. Kuhfuss. Overview of the I-Way: Wide area visual supercomputing. International Journal of Supercomputer Applications, 10(2), 1996.Google Scholar
- 4.Ian Foster, Jonathan Geisler, Carl Kesselman, and Steven Tuecke. Managing multiple communication methods in high-performance networked computing systems. Parallel and Distributed Computing, 1996.Google Scholar
- 5.A. Geist, A. Beguelin, J. Dongarra, W. Liang, B. Manchek, and V. Sunderam. PVM: Parallel Virtual machine — A User's Guide and Tutorial for Network Parallel Computing. MIT Press, 1994.Google Scholar
- 6.GENIAS Software GmbH. Codine: Computing in distributed networked environments. URL http://www.genias.de/genias/english/codine.html, September 1995.Google Scholar
- 7.W. Gropp and E. Lusk. MPICH working note: Creating a new MPICH device using the channel interface. Technical Report, Mathematics and Computer Science Division, Argonne National Laboratory, 1995.Google Scholar
- 8.R. Hempel. The ANL/GMD macros (PARMACS) in Fortran for portable parallel programming using the message passing programming modell — User's Guide and Reference Manual. Technical Report, GMD, 1991.Google Scholar
- 9.Message Passing Interface Forum. MPI: A message-passing interface standard. International Journal of Supercomputer Applications, 8(3/4), 1994.Google Scholar
- 10.H. Mierendorff, K. Stueben, C.-A. Thole, and O. Thomas. Europort-1: Porting industrial codes to parallel architectures. Springer LNCS 919, 1995.Google Scholar
- 11.J. Pruyne and M. Livny. Parallel processing on dynamic resources with CARMI. Lecture Notes in Computer Science, 1995.Google Scholar
- 12.F. Ramme, T. Römke, and K. Kremer. A distributed computing center software for efficient use of parallel computer systems. In High-Performance Computing and Networking, volume 2 of Springer LNCS 797 Vol. II, pages 129–136, 1994.Google Scholar
- 13.A. Reinefeld, J. Gehring, M. Brune. Linking Communication Models in Heterogeneous Computing Networks. To appear in Journal of Systems Architecture, Elsevier 1997Google Scholar
- 14.W. Richard Stevens. UNIX Network Programming. Prentice Hall, 1990.Google Scholar
- 15.R. van Renesse, K. Birman, R. Friedmann, M. Hayden, and D. Karr. A framework for protocol composition in Horus. In Proc. Principles of Distributed Computing Conf., 1995.Google Scholar