[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ale] high performance computing

Subject: [ale] high performance computing
From: laytonjb at att.net (Jeff Layton)
Date: Thu, 26 Jul 2012 14:45:21 -0400
In-reply-to: <CAEo=5PzUi8PTn2K6qKKn92o1Q0N3OMO5TsPwg7RQXiSJ15HLGQ@mail.gmail.com>
References: <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <CAEo=5PzUi8PTn2K6qKKn92o1Q0N3OMO5TsPwg7RQXiSJ15HLGQ@mail.gmail.com>

  On 07/26/2012 02:33 PM, Jim Kinney wrote:
> FFT with CUDA?? All the FFT code I've seen is floating point which
> CUDA can't do.

Sure. It's very common and comes with CUDA.

CuFFT:
http://developer.download.nvidia.com/compute/DevZone/docs/html/CUDALibraries/doc/CUFFT_Library.pdf
http://developer.download.nvidia.com/GTC/PDF/GTC2012/PresentationPDF/S0209-GTC2012-3D-FFT-CUDA-4.pdf
http://developer.download.nvidia.com/compute/cuda/2_2/sdk/website/projects/convolutionFFT2D/doc/convolutionFFT2D.pdf

https://sites.google.com/site/zilong308/cuda/cuda-fft-library-cufft
http://blogs.hoopoe-cloud.com/index.php/2009/01/using-cuda-fft-from-fortran/

Jeff

Follow-Ups:
- [ale] high performance computing
  - From: ecashin at noserose.net (Ed Cashin)

References:
- [ale] high performance computing
  - From: john at johnheim.net (John Heim)
- [ale] high performance computing
  - From: vernard at venger.net (Vernard Martin)
- [ale] high performance computing
  - From: laytonjb at att.net (Jeff Layton)
- [ale] high performance computing
  - From: john at johnheim.net (John Heim)
- [ale] high performance computing
  - From: laytonjb at att.net (Jeff Layton)
- [ale] high performance computing
  - From: jim.kinney at gmail.com (Jim Kinney)

Prev by Date: [ale] high performance computing
Next by Date: [ale] DISA considering git for Forge.mil
Previous by thread: [ale] high performance computing
Next by thread: [ale] high performance computing
Index(es):
- Date
- Thread