It is important to vet your GTX cards with the gpu validation test for
system sizes comparable to the largest system or combinations of runs you
plan to execute. You should do this on a regular basis too, as cards age
and experience power events. We have had been lucky with EVGA cards from
Newegg, but using a vendor like Exxact to sell you a pre-vetted card or
system is a lot simpler, and I prefer that now. I would look at a vendor
willing to vet your cards with AMBER and I'd be surprised if Dell would do
that. I have a TitanXP from Exxact that can run two simulations with
reasonable performance due to the large memory. But, I vetted it
beforehand for that particular amount of required memory. We have had
cards that tested perfectly after initial installation, but a year or more
later, starting testing with failures above certain required amounts of
memory. Maybe having a regular schedule testing every 6 months for
beginning failures starting with the largest system your card can handle
would be wise. I'd certainly hate to have months of MD be in question due
to faulty calculations.
I got an email from Exxact today advertising academic discounts on Volta
cards. I haven't gotten a price yet, but certainly if AMBER could support
CUDA 9 and the Volta architecture, we would be figuring out a way to
acquire at least one this Fall.
Sincerely,
Dow
⚛Dow Hurst, Research Scientist
340 Sullivan Science Bldg.
Dept. of Chem. and Biochem.
University of North Carolina at Greensboro
PO Box 26170 Greensboro, NC 27402-6170
On Mon, Oct 30, 2017 at 5:29 PM, Gerald Monard <
Gerald.Monard.univ-lorraine.fr> wrote:
> Hello,
>
> Just to add up to Ross' warning.
>
> I've been discussing with Dell Computers in France for the recent
> months. In our HPC centre, we want to buy Dell computers with GTX cards
> to "attract" people who, most notably, run Amber, Gromacs, etc. on GPUs.
> We also have NVIDIA's P100 with NVLink but we would like to encourage
> people who don't need such double precision power (thanks for the SPFP
> versioN!) to use cheaper and as efficient GTX cards.
>
> Three information:
> 1) Dell does not officially support GTX cards on their HPC servers (and
> has never had), but so far, you could buy them separately and plug them in.
> 2) with the new Dell generation (14G), it won't be even possible to plug
> GTX cards in these servers (for those who are familiar with Dell
> portfolio, I'm talking about R740 and T640). From what I've been told,
> this is at NVIDIA's request. We are currently trying to buy some servers
> from the previous generation (13G) while they are still available to be
> able to plug whatever GTX card we could buy, but ("don't know why") my
> Dell representatives are very slow to provide quotations.
> 3) At a recent Dell-Intel event in Europe, some Dell representative asks
> the audience (mostly made from people in the academic HPC business) how
> they feel about Dell portfolio. Very first comment/question from the
> audience: "when will you finally support GTX cards in your server? They
> are cheap, works fine for many purpose, professional cards are too
> expensive", then all the audience nodded and said yes this is what we
> would like to see. The Dell representative didn't (couldn't?) answer us.
>
> I do agree with Ross' opinion, but the AI marked being so strong as the
> moment and generating so much money for NVIDIA, I don't really know what
> we can do about it.
>
> Other comment: in my (small) university, some lab has started to buy GTX
> cards by the dozens, especially those working in the molecular dynamics
> fields (but not only). People know how terrific the numbers are with
> Amber, Gromacs, etc and I guess that NVIDIA has started to realize that
> a significant portion of its high-end GTX sales were not going to
> gamers... .Ross, Scott et al.: your work on porting pmemd to cheap gpu
> cards has been too terrific!
>
> Gerald.
>
> On 10/30/2017 07:57 PM, Ross Walker wrote:
> > Dear All,
> >
> > In the spirit of open discussion I want to bring the AMBER community's
> attention to a concern raised in two recent news articles:
> >
> > 1) "Nvidia halts distribution partners from selling GeForce graphics
> cards to server, HPC sectors" - http://www.digitimes.com/news/
> a20171027PD200.html
> >
> > 2) "Nvidia is cracking down on servers powered by Geforce graphics
> cards" - https://www.pcgamesn.com/nvidia-geforce-server
> >
> > I know many of you have benefitted greatly over the years from the GPU
> revolution that has transformed the field of Molecular Dynamics. A lot of
> the work in this field was provided by people volunteering their time and
> grew out of the idea that many of us could not have access to or could not
> afford supercomputers for MD. The underlying drive was to bring
> supercomputing performance to the 99% and thus greatly extend the amount
> and quality of science each of us could do. For AMBER this meant supporting
> all three models of NVIDIA graphics card, Geforce, Quadro and Tesla in
> whatever format or combination, you the scientist and customer, wanted.
> >
> > In my opinion key to AMBER's success was the idea that, for running MD
> simulations, very few people in the academic field, and indeed many R&D
> groups within companies, small or large, could afford the high end tesla
> systems, whose price has been steadily going up substantially above
> inflation with each new generation (for example the $149K DGX-1). The
> understanding, both mine and that of the field in general, has essentially
> always been that assuming one was willing to accept the risks on
> reliability etc, use of Geforce cards should be perfectly reasonable. We
> are not after all running US air traffic control, or some other equally
> critical system. It is prudent use of limited R&D funds, or in many cases
> tax payer money and we are the customers after all so should be free to
> choose the hardware we buy. NVIDIA has fought a number of us for many years
> on this front but mostly in a passive aggressive stance with the occasional
> personal insult or threat. As highlighted in the abo
> > ve artic
> > les with the latest AI bubble they have cemented a worrying monopoly
> and are now getting substantially more aggressive, using this monopoly to
> pressure suppliers to try to effectively ban the use of Geforce cards for
> scientific compute and restrict what we can buy to Tesla cards, that for
> the vast majority of us are simply out of our price range.
> >
> > In my opinion this a very worrying trend that could hurt us all and have
> serious repercussions on all of our scientific productivities and the field
> in general. If this is a concern to you too I would encourage each of you
> to speak up. Contact people you know at NVIDIA and make your concerns
> heard. I am concerned that if we as a community do not speak up now we
> could see our field be completely priced out of the ability to make use of
> GPUs for MD over the next year.
> >
> > All the best
> > Ross
> >
> >
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
>
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Oct 30 2017 - 16:30:02 PDT
