Multigrid on massively parallel architectures

Conference · Fri Sep 17 00:00:00 EDT 1999

DOI:https://doi.org/10.1007/978-3-642-58312-4_13· OSTI ID:13917

Falgout, R D; Jones, J E

The scalable implementation of multigrid methods for machines with several thousands of processors is investigated. Parallel performance models are presented for three different structured-grid multigrid algorithms, and a description is given of how these models can be used to guide implementation. Potential pitfalls are illustrated when moving from moderate-sized parallelism to large-scale parallelism, and results are given from existing multigrid codes to support the discussion. Finally, the use of mixed programming models is investigated for multigrid codes on clusters of SMPs.

• Research Organization: Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
• Sponsoring Organization: USDOE Office of Defense Programs (DP) (US)
• DOE Contract Number: W-7405-ENG-48
• OSTI ID: 13917
• Report Number(s): UCRL-JC-133948; DP0102051; DP0102051; TRN: AH200135%%492
• Resource Relation: Journal Volume: 14; Conference: Sixth European Multigrid Conference, Gent (BE), 09/27/1999--09/30/1999; Other Information: PBD: 17 Sep 1999
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
ALGORITHMS
IMPLEMENTATION
PERFORMANCE
PARALLEL PROCESSING