Performance measurements of the 3d FFT on the Blue Gene/L supercomputer
| Venue: | Euro-Par 2005 Parallel Processing: 11th International Euro-Par Conference |
| Citations: | 3 - 3 self |
BibTeX
@INPROCEEDINGS{Eleftheriou_performancemeasurements,
author = {Maria Eleftheriou and Blake Fitch and R Rayshubskiy and T. J. Christopher Ward and Robert Germain},
title = {Performance measurements of the 3d FFT on the Blue Gene/L supercomputer},
booktitle = {Euro-Par 2005 Parallel Processing: 11th International Euro-Par Conference},
year = {},
pages = {795--803},
publisher = {Springer-Verlag}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
Abstract. This paper presents performance characteristics of a communicationsintensive kernel, the complex data 3D FFT, running on the Blue Gene/L architecture. Two implementations of the volumetric FFT algorithm were characterized, one built on the MPI library using an optimized collective all-to-all operation [2] and another built on a low-level System Programming Interface (SPI) of the Blue Gene/L Advanced Diagnostics Environment (BG/L ADE) [17]. We compare the current results to those obtained using a reference MPI implementation (MPICH2 ported to BG/L with unoptimized collectives) and to a port of version 2.1.5 the FFTW library [14]. Performance experiments on the Blue Gene/L prototype indicate that both of our implementations scale well and the current MPI-based implementation shows a speedup of 730 on 2048 nodes for 3D FFTs of size 128 × 128 × 128. Moreover, the volumetric FFT outperforms FFTW port by a factor 8 for a 128 × 128 × 128 complex FFT on 2048 nodes. 1
