charm AT lists.cs.illinois.edu

Subject: Charm++ parallel programming system

List archive

Re: [charm] [ppl] Distributed FFT in CHARM++

From: Phil Miller <mille121 AT illinois.edu>
To: Benson Muite <muite AT umich.edu>
Cc: Charm Mailing List <charm AT cs.illinois.edu>
Subject: Re: [charm] [ppl] Distributed FFT in CHARM++
Date: Wed, 5 Sep 2012 09:45:36 -0500
List-archive: <http://lists.cs.uiuc.edu/pipermail/charm/>
List-id: CHARM parallel programming system <charm.cs.uiuc.edu>

On Sat, Aug 25, 2012 at 4:23 PM, Benson Muite
<muite AT umich.edu>
wrote:
> Could you please add documentation on how global FFT is implemented in
> Charm++ in your online tutorial.
>
> Performance information on this would also be useful. If this
> information is already available, I would appreciate knowing where I
> should look.

A simple implementation of a distributed transpose-based 1D FFT can be found
at
https://charm.cs.illinois.edu/cgi-bin/gitweb2.cgi?p=benchmarks/fft-trans.git;a=summary
git://charm.cs.illinois.edu/benchmarks/fft-trans.git

Performance results from this implementation appear in our HPC
challenge submission report from SC11, which can be found at
http://charm.cs.illinois.edu/papers/11-49

Hope this helps!

Phil

Re: [charm] [ppl] Distributed FFT in CHARM++, Phil Miller, 09/05/2012