Home >

news Help

Publication Information


Title
Japanese: 
English:Bandwidth intensive 3-D FFT kernel for GPUs using CUDA 
Author
Japanese: 額田彰, 尾形泰彦, 遠藤敏夫, 松岡聡.  
English: Akira Nukada, Yasuhiko Ogata, Toshio Endo, Satoshi Matsuoka.  
Language English 
Journal/Book name
Japanese: 
English:Proceedings of the 2008 ACM/IEEE conference on Supercomputing (SC08) 
Volume, Number, Page         pp. 1-11
Published date Nov. 2008 
Publisher
Japanese: 
English:IEEE 
Conference name
Japanese: 
English:2008 ACM/IEEE conference on Supercomputing (SC08) 
Conference site
Japanese: 
English:Austin, Texas 
DOI https://doi.org/10.1109/SC.2008.5213210
Abstract Most GPU performance "hypes" have focused around tightly-coupled applications with small memory bandwidth requirements e.g., N-body, but GPUs are also commodity vector machines sporting substantial memory bandwidth; however, effective programming methodologies thereof have been poorly studied. Our new 3-D FFT kernel, written in NVIDIA CUDA, achieves nearly 80 GFLOPS on a top-end GPU, being more than three times faster than any existing FFT implementations on GPUs including CUFFT. Careful programming techniques are employed to fully exploit modern GPU hardware characteristics while overcoming their limitations, including on-chip shared memory utilization, optimizing the number of threads and registers through appropriate localization, and avoiding low-speed stride memory accesses. Our kernel applied to real applications achieves orders of magnitude boost in power&cost vs. performance metrics. The off-card bandwidth limitation is still an issue, which could be alleviated somewhat with application kernels confinement within the card, while ideal solution being facilitation of faster GPU interfaces.

©2007 Institute of Science Tokyo All rights reserved.