کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
503454 | 863768 | 2007 | 8 صفحه PDF | دانلود رایگان |
QCDOC is a massively parallel supercomputer with tens of thousands of nodes distributed on a six-dimensional torus network. The 6D structure of the network provides the needed communication resources for many communication-intensive applications. In this paper, we present a parallel algorithm for three-dimensional Fast Fourier Transform and its implementation for a 4096-node QCDOC prototype. Two techniques have been used to increase its parallel performance: simultaneous multi-dimensional communication and communication-and-computation overlapping. Benchmarking experiments suggest that 3D FFTs of size 128×128×128128×128×128 can scale well on such platforms up to 4096 nodes. Our performance results suggest stronger scalability on QCDOC than on IBM BlueGene/L supercomputer.
Journal: Computer Physics Communications - Volume 176, Issue 8, 15 April 2007, Pages 531–538