Scalability issues for a class of CFD applications

Author

Naik, Vijay K.

Author_Institution

IBM, Thomas J. Watson Res. Center, Yorktown Heights, NY, USA

fYear

1992

fDate

26-29 Apr 1992

Firstpage

268

Lastpage

275

Abstract

Considers the performance scalability of a class of computational fluid dynamics applications. The results indicate that neither the scalability in time nor the scalability in problem size can be obtained by simply scaling up the processing power. Results are presented to show that latency, packet size, and transmission speeds play an important role. However, improvements only in the architectural parameters are not sufficient to realize full performance scalability. Suitable partitioning and algorithmic parameters must be selected for each type of architecture

Keywords

fluid dynamics; mechanical engineering computing; parallel algorithms; parallel architectures; performance evaluation; physics computing; algorithmic parameters; architectural parameters; computational fluid dynamics applications; latency; packet size; partitioning parameters; performance scalability; transmission speeds; Computational fluid dynamics; Computational geometry; Costs; Delay; Parallel machines; Parallel processing; Partitioning algorithms; Physics computing; Processor scheduling; Scalability;

fLanguage

English

Publisher

ieee

Conference_Titel

Scalable High Performance Computing Conference, 1992. SHPCC-92, Proceedings.

Conference_Location

Williamsburg, VA

Print_ISBN

0-8186-2775-1

Type

conf

DOI

10.1109/SHPCC.1992.232632

Filename

232632