[Beowulf] AMD performance (was 500GB systems)

Stu Midgley sdm900 at gmail.com
Mon Jan 14 15:38:16 PST 2013


We will use the phi's in many ways.  By having codes directly compiled on
them and running directly on the phi (ssh to the phi and feed data on
stdin)...  some of our codes are trivially threads with pthreads and some
with openmp.  We also have a GPU code (opencl) which will hopefully work
out of the box as well.


On Tue, Jan 15, 2013 at 12:02 AM, Alan Louis Scheinine <
alscheinine at tuffmail.us> wrote:

> Stu Midgley wrote:
>
>> As you might guess, we were very happy with how our codes run on the Phi
>>
> > and the time/effort required to port.  It is very very simple to use and
> > the performance is excellent :)  With no tuning (just recompile) we saw
> > a single phi go at about 1.7x faster than our current AMD 64 cores nodes.
>
> What parallel language for utilization of Phi?  OpenMP?
>
> Regards,
> Alan Scheinine
>
> --
>
>  Alan Scheinine
>  200 Georgann Dr., Apt. E6
>  Vicksburg, MS  39180
>
>  Email: alscheinine at tuffmail.us
>  Mobile phone: 225 288 4176
>
>  http://www.flickr.com/photos/**ascheinine<http://www.flickr.com/photos/ascheinine>
>



-- 
Dr Stuart Midgley
sdm900 at sdm900.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.beowulf.org/pipermail/beowulf/attachments/20130115/303eadf0/attachment.html 


More information about the Beowulf mailing list