Name
Abstract | ||
|---|---|---|
In this paper, we present our implementation of the fast Fourier transforms on graphic processing unit (GPU) using OpenCL. This implementation of the FFT (ToPe-FFT) is based on the Cooley-Tukey set of algorithms with support for 1D and higher dimensional transforms using different radices. Factorization for mix-radices enables our code to target FFTs of near arbitrary length. In systems with multiple graphic cards (GPUs), the library automatically balances the FFT computation thus achieving maximum resource utilization and higher speedup. Based on profiling and micro-benchmarking of ToPe-FFT, it is observed that the average speedup of our library for different sizes is 48x faster than the single CPU-based code using FFTW and 3x faster than NVIDIA's GPU-based cuFFT library. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1002/cpe.4256 | CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE |
Keywords | Field | DocType |
fast fourier transform,GPU,high performance computing,OpenCL | Supercomputer,Computer science,Profiling (computer programming),Parallel computing,Software,Fast Fourier transform,Factorization,Speedup,Computation | Journal |
Volume | Issue | ISSN |
29 | 21 | 1532-0626 |
Citations | PageRank | References |
2 | 0.37 | 3 |
Authors | ||
6 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Bilal Jan | 1 | 22 | 2.45 |
| Fiaz Gul Khan | 2 | 45 | 5.12 |
| Bartolomeo Montrucchio | 3 | 181 | 28.42 |
| Anthony T. Chronopoulos | 4 | 523 | 50.61 |
| Shahaboddin Shamshirband | 5 | 512 | 53.36 |
| Abdul Nasir Khan | 6 | 257 | 14.85 |