[Q-e-developers] Call of GPU benchmark inputs for PWSCF

[Lorenzo Paulatto]

I'm attaching an input for graphene which takes about 5-10 minutes to run.
It has a very large number of k-points and large cutoff, but only 4 atoms.

The pseudopotential is here:  
http://quantum-espresso.org/pseudo/1.3/UPF/C.pz-vbc.UPF

cheers

On Sat, 11 Feb 2012 15:41:44 +0100, Filippo Spiga <filippo.spiga at ichec.ie>
wrote:

> Dear all,
>     the GPU porting of PWSCF carried in the past months is going to hit  
> an important milestone: the GPU Technology Conference in San Jose (USA).  
> For that date, we would like to deliver the full package and finally  
> merge the GPU branch into the main code.
>
> What it is really important in a project like this is a detailed and a  
> 365-degree assessment of the performance, taking into account as many  
> different scenario (read as many different type of input/calculation) as  
> possible. In the next (missing) two months we are going to finalize some  
> important piece of code and run tests. A lot of tests. As much tests as  
> we can.
>
> The purpose of this email is to collect from users new benchmark inputs  
> that reflect real scientific challenges. We are looking to new input for  
> serial and parallel calculations, small or big (not very big... let's  
> say "medium") systems, gamma or k-point, with or without magnetization,  
> with or without exact-exchange, etc...
>
> What we are especially looking for "special" inputs. I mean, inputs that  
> are able to intensively stress some specific (and well localized) part  
> of the code.
>
> Do not hesitate to contact me. I look forward to receiving new test  
> cases!
> At the end it will be a pleasure to officially acknowledge all the  
> people that will contribute.
>
> Regards,
> Filippo
>


-- 
Lorenzo Paulatto IdR @ IMPMC/UPMC CNRS & Université Paris 6
phone: +33 (0)1 44275 084 / skype: paulatz
www:   http://www.impmc.upmc.fr/~paulatto/
mail:  23-24/4é14 Boîte courrier 115, 4 place Jussieu 75252 Paris Cédex 05
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gite.in
Type: application/octet-stream
Size: 635 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20120211/4938aa53/attachment.obj>