The same-source parallel MM{sub 5} [electronic resource].
- Published:
- Washington, D.C. : United States. Dept. of Energy, 1999. and Oak Ridge, Tenn. : Distributed by the Office of Scientific and Technical Information, U.S. Dept. of Energy.
- Physical Description:
- 7 pages : digital, PDF file
- Additional Creators:
- Argonne National Laboratory, United States. Department of Energy, and United States. Department of Energy. Office of Scientific and Technical Information
- Access Online:
- www.osti.gov
- Restrictions on Access:
- Free-to-read Unrestricted online access
- Summary:
- The set of architectures available to users of the Penn State/NCAR MM5 has been expanded to included distributed-memory parallel computers, providing cost-effective scalable performance and memory capacity for large problem sizes. The same-source approach uses high-level parallel library and source-translation technology for adapting MM5, simplifying maintenance and allowing new physics modules to be incorporated without modification. The approach facilitates maintenance of the DM-parallel option to MM5 as an option within the official version, rather than as a separate stand-alone version. As a result, the DM-parallel option to MM5 (now at Version 3.1) has been a part of six subsequent model releases since MM5 Version 2.8 in March 1998. The same-source approach is applicable to other, similarly constructed codes when there is a need or desire to develop the code for distributed memory parallel machines without impacting the pre-existing source code. The approach is also compatible with pre-existing loop-level multithreading directives so that the code will run in distributed-memory/shared-memory mode on SMP clusters.
- Subject(s):
- Note:
- Published through SciTech Connect., 08/23/1999., "anl/mcs/cp-99863", 2nd International Workshop on Software Engineering and Code Design in Parallel Meteorological and Oceanographic Applications, Scottsdale, AZ (US), 06/09/1998--06/12/1998., and Michalakes, J.
- Funding Information:
- W-31109-ENG-38
View MARC record | catkey: 14136723