High-performance parallel FFT algorithms for the HITACHI SR8000
Daisuke Takahashi · 2000
We propose high-performance parallel one-dimensional fast Fourier transform (FFT) algorithms for distributed memory parallel computers with vector symmetric multiprocessor (SMP) nodes. The four-step FFT algorithm can be altered into a five-step FFT algorithm to expand the innermost loop length. We use the four-step and five-step algorithms to implement the parallel one-dimensional FFT algorithms. In our proposed parallel FFT algorithms, since we use cyclic distribution, all-to-all communication takes place only once. Moreover, the input data and output data are both in natural order. Performance results of one-dimensional power-of-two FFTs on a distributed memory parallel computer with (pseudo) vector SMP nodes, HITACHI SR8000, are reported. We succeeded in obtaining performance of about 38 GFLOPS on a 16-node SR8000.