[gmx-developers] Current status FFTW-3D

[Erik Lindahl]

Hi Roland,

On Aug 30, 2007, at 6:39 PM, Roland Schulz wrote:
>>
> I did a performance test on Jaguar (2.6 GHz dual-core Opteron, Cray
> Seastar router) which is attached. I hope attachments are fine on this
> list. I did it only on this cluster because I wanted to compare it to
> the Eleftheriou paper and thus till 2048 and this is the only of this
> size I have access to at the moment. I think it looks good in
> comparison, because I didn't do any special optimization. I may be  
> able
> to do some performance tests on BlueGene to be able to do a direct
> comparison.

That looks great as a starting point. Be aware that we probably want  
to use real-to-complex transforms though, since that cuts the  
executing time in half (and, more important, the communication  
bandwidth requirements).
>
>> - In the end we would like to have the copyright of new code, but  
>> that
>> is not of immediate concern, I guess we first need a working code.
> It is GPL so it shouldn't be a problem.

That is fine if we make it a separate (optional) library. The reason  
for this is that a couple of projects such as Folding at Home have our  
permission to ship non-GPL binaries (that create encrypted  
trajectories so users can't manipulate the science), and if we  
include GPL-only code we don't have (shared) copyright or free  
distribution rights to we wouldn't be able to make any exceptions at  
all.

Cheers,

Erik
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20070831/39707ad8/attachment.html>