Title :
Compute intensity and the FFT
Author_Institution :
Cray Res. Superservers, Inc., Beaverton, OR, USA
Abstract :
This paper describes how high compute intensity programming techniques combined with algorithms in the literature can result in efficient single- and multi-dimensional FFTs on large numbers of processors on the CRAY APP. The CRAY APP is a shared-memory parallel computer based on the Intel i860 microprocessor. It incorporates up to 84 i860s in an architecture which allows for very efficient gang scheduling and barrier synchronization. FFT performance figures for various data set sizes and processor configurations are included.
Keywords :
computational complexity; fast Fourier transforms; parallel algorithms; parallel programming; processor scheduling; software performance evaluation; CRAY APP; FFT; Intel i860 microprocessor; barrier synchronization; data set sizes; fast Fourier transforms; gang scheduling; high compute intensity programming; parallel algorithms; performance figures; processor configurations; shared-memory parallel computer; Bandwidth; Computer architecture; Concurrent computing; Fast Fourier transforms; Flexible printed circuits; High performance computing; Memory management; Microprocessors; Parallel architectures; Supercomputers;
Conference_Titel :
Supercomputing '93. Proceedings
Print_ISBN :
0-8186-4340-4
DOI :
10.1109/SUPERC.1993.1263522
