A Comparison of Distributed Memory Fast Fourier Transform (FFT) Library Packages
In this BoF developers and users of different FFT packages will answer the following questions in short presentations that lasts not more than 10 min.
- Why did you write your own FFT?
- What considerations are important for you in an FFT implementation?
- What might you look for if there were to be a unified FFT interface (similar to BLAS, LAPACK and SCALAPACK interfaces)?
- How important are performance, portability, and scalability for you?
- Will FFT be needed in exascale computing and if so how will it be achieved?
- What would be a good FFT benchmark or a good way to include the FFT in a high-performance computer benchmark?
Following these questions, there will be a general discussion between the presenters along with audience interaction. Contributions will highlight existing or potential bottlenecks in the different FFT packages; possible methods to overcome them as well as FFT interface design.
|Daisuke Takahashi||University of Tsukuba, Japan||FFTE||Slides|
|Franz Franchetti ,Thom Popovoci||Carnegie Mellon University, USA||Spiral||Slides|
|Yu Feng||Berkeley Center for Cosmological Physics, USA||Pfft.py||Slides|
|Jens Henrik Gobbert||Julich Supercomputing Center, Germany||nb3dfft||Slides|
|Sunita Chandrasekaran||University of Delaware, USA||PsFFT, cusFFT||Slides|
|Amir Gholami||University of California at Berkeley, USA||AccFFT|
|Dmitry Pekurovsky||San Diego Supercomputing Center, USA||P3DFFT||Slides|
|Steve Plimpton||Sandia National Laboratories, USA||Parallel FFT||Slides|