[Users] What/where is thorn 'DGEMM'?

Erik Schnetter schnetter at cct.lsu.edu
Sun Jun 19 22:06:28 CDT 2011


I believe thorn DGEMM is my fault: This is a thorn I wrote at some
point to measure the raw floating point performance of the CPUs in a
simulation, to detect cases where this performance is suddenly much
lower than expected. This thorn is available in the LSUThorns
repository. However, you don't need it -- it should not be activated
in an example parameter file.

I just see that the url <https://svn.cct.lsu.edu/> list all subversion
repositories at the CCT, including DGEMM... but that doesn't help you
run examples.

I'm afraid we don't really have a good set of examples (we should!).
We used to have a few with the first version of the Einstein Toolkit,
but these are by now probably out of date. Your best bet is probably
to ask for a good parameter file e.g. for a binary black hole
collision or a neutron star calculation, and then to simplify/modify
this until it suits your needs.

-erik

On Sun, Jun 19, 2011 at 10:22 PM, Scott Hawley <scott.hawley at belmont.edu> wrote:
> I'm trying to run...any.. par file that came with the ETK checkout.  Some of them require thorns not included in the distribution, e.g thorn PsiKadelia needed by many par files in EinsteinInitialData.
>
> The files in CarpetRegrid2/par/ require a thorn called DGEMM, which isnt part of the ETK list and doesn't even appear in Google:
>
> "Sorry, Cactus dgemm thorn did not return any results."
>
> How does one run, for example, staticbh.par?
>
> Thanks
>
> %Sent from my iPhone
> _______________________________________________
> Users mailing list
> Users at einsteintoolkit.org
> http://lists.einsteintoolkit.org/mailman/listinfo/users
>



-- 
Erik Schnetter <schnetter at cct.lsu.edu>   http://www.cct.lsu.edu/~eschnett/


More information about the Users mailing list