<div dir="ltr"><br><div>You have to talk to the right people at NVIDIA ... benchmarking group.</div><div><br></div><div>The version I am using from 2018 is:</div><div><br></div><div> xhpl_cuda9.2.88_mkl_2018_ompi_3.1.0_gcc485_sm35_sm60_sm70_5_
18_18</div><div><br></div><div>but there must be something more current than that now. This one works out<br></div><div>through the V100 as the name implies.</div><div><br></div><div>rbw</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Wed, Aug 14, 2019 at 12:31 PM Michael Di Domenico <<a href="mailto:mdidomenico4@gmail.com">mdidomenico4@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">yeah, i'm not surprised. the original developer did it as part of a<br>
research project i believe. it was never updated after he wrote the<br>
paper and published.<br>
<br>
i wish nvidia would release the hpcg benchmark source code. all you<br>
can get is the precompiled binaries which i can't use. i've asked my<br>
contacts, but i guess i don't have enough sway. :(<br>
<br>
On Wed, Aug 14, 2019 at 1:12 PM Prentice Bisbal via Beowulf<br>
<<a href="mailto:beowulf@beowulf.org" target="_blank">beowulf@beowulf.org</a>> wrote:<br>
><br>
> I looked into this further, and the version available is for the Fermi<br>
> GPUs, and doesn't really work with the V100 GPUs we have in our target<br>
> system. :(<br>
><br>
> --<br>
> Prentice<br>
><br>
> On 8/14/19 11:17 AM, Michael Di Domenico wrote:<br>
> > gpu linpack from nvidia is available via the developer portal. but<br>
> > you can probably also reach out to the developer directly, he's<br>
> > friendly.<br>
> ><br>
> > though they haven't updated it in a long time, not sure if it still<br>
> > runs on the newer cards.<br>
> ><br>
> > On Wed, Aug 14, 2019 at 11:15 AM Prentice Bisbal via Beowulf<br>
> > <<a href="mailto:beowulf@beowulf.org" target="_blank">beowulf@beowulf.org</a>> wrote:<br>
> >> I have a new GPU-based cluster that I'd like to bench mark with<br>
> >> High-Performance LINPACK (HPL). Does anyone know where I can get a<br>
> >> version of HPL written for GPUs? I know NVIDIA has a version, and I've<br>
> >> already reached out to NVIDIA to see if they'll share their version with<br>
> >> me. I just figured I'd ask here, in case there are other sources.<br>
> >><br>
> >> --<br>
> >> Prentice<br>
> >><br>
> >> _______________________________________________<br>
> >> Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org" target="_blank">Beowulf@beowulf.org</a> sponsored by Penguin Computing<br>
> >> To change your subscription (digest mode or unsubscribe) visit <a href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf" rel="noreferrer" target="_blank">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a><br>
> > _______________________________________________<br>
> > Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org" target="_blank">Beowulf@beowulf.org</a> sponsored by Penguin Computing<br>
> > To change your subscription (digest mode or unsubscribe) visit <a href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf" rel="noreferrer" target="_blank">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a><br>
> _______________________________________________<br>
> Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org" target="_blank">Beowulf@beowulf.org</a> sponsored by Penguin Computing<br>
> To change your subscription (digest mode or unsubscribe) visit <a href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf" rel="noreferrer" target="_blank">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a><br>
_______________________________________________<br>
Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org" target="_blank">Beowulf@beowulf.org</a> sponsored by Penguin Computing<br>
To change your subscription (digest mode or unsubscribe) visit <a href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf" rel="noreferrer" target="_blank">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a><br>
</blockquote></div>