Distributed processing of a regional prediction model

K. W. Johnson, J. Bauer, G. A. Riccardi, K. K. Droegemeier, Ming Xue

Research output: Contribution to journalArticlepeer-review


A domain decomposition strategy using a network of workstations produced a significant decrease in elapsed time and increase in problem size relative to a single-workstation run. The peformance of the resulting program is described by derived formulas (collectively known as a performance model), which predict the execution time and speedup for different numbers of processors and problem sizes. The interprocessor communication speeds are shown to be the major obstacle to achieving full processor use. The effect of faster communication networks on parallel performance is predicted based on this performance model. Parallelization experiments using the ARPS code were run on a cluster of IBM RS6000 workstations connected via Ethernet. -from Authors

Original languageEnglish (US)
Pages (from-to)2558-2572
Number of pages15
JournalMonthly Weather Review
Issue number11
StatePublished - 1994
Externally publishedYes

ASJC Scopus subject areas

  • Atmospheric Science


Dive into the research topics of 'Distributed processing of a regional prediction model'. Together they form a unique fingerprint.

Cite this