Parallel Sparse and Conventional FFTs, Applications and Implementation
The fast Fourier Transform (FFT) is an algorithm used in a wide variety of applications, yet does not make optimal use of many current and emerging platforms such as many-core processors, GPUs, and distributed-memory systems. Hardware utilization performance on its own does not, however, imply optimal problem-solving. The purpose of this minisymposium is to enable an exchange of information between people working on FFT algorithms such as sparse and conventional FFTs, to those working on FFT implementations, in particular for parallel hardware.
| Speaker | Affiliation | Presentation |
|---|---|---|
| Franz Fanchetti | Carnegie Mellon University, USA | Slides |
| Andrew Canning | Lawrence Berkeley National Laboratory, USA | |
| Dmitry Pekurovsky | University of California San Diego, USA | Slides |
| Daisuke Takahashi | University of Tsukuba, Japan | Slides |
| Samar Aseeri | King Abdullah University of Science and Technology (KAUST), Saudi Arabia | Slides |
| Lisandro Dalcin | King Abdullah University of Science and Technology (KAUST), Saudi Arabia | Slides |
| John Bowman | University of Alberta, Canada | Slides |
| Malcolm Roberts | AMD, Canada | Slides |