Re: [AMBER] Possible concerns regarding future availability of cost effective NVIDIA GPU systems for running AMBER

From: David Cerutti <>
Date: Fri, 8 Dec 2017 15:18:39 -0500

I was just discussing Titan-V with one of my colleagues upstairs. That's
really incredible--I thought the reason people paid $6k for a GP100 was the
increase in double precision capability, and perhaps some better assurances
of consistency in the calculations. But if Titan-V can do in fp64 all that
the Volta we're presently looking at does in fp32, I wonder if we shouldn't
consider a new compilation mode. There are some things about DPFP
regarding memory access that are way sub-optimal, so it's not just slower
because there are fewer fp64 registers. Also, the amount of __shared__
probably isn't doubling in concert with the larger number of 64-bit
registers. But, if we can do more fp64, a high precision computation,
fixed precision accumulation (HPFP) mode might be appealing.


On Fri, Dec 8, 2017 at 2:34 PM, Dow Hurst <> wrote:

> I think the new Nvidia Titan V just announced
> <
> its-monster-volta-gpu-to-a-graphics-cards-and-it-costs-3000/>
> would have some positive impact on the possibility of the development of
> Amber on GPUs for the Volta architecture. Seems that a $3K card, even
> though expensive, is much more in reach than the original $10K card. The
> price will drop as time passes making it even more accessible. What kind of
> improvements to Amber could be made within the scope of the Titan V? I
> understand it won't have any SLI connections and is still focused on
> computation rather than gaming. Would Scott Le Grand's comment still hold
> true?
> "*What NVIDIA will miss out on is beating Anton with one or two DGX-1Vs.
> That's something I think could be achieved in a matter of months, but it
> requires rewriting the inner loops from the ground up to parallelize
> everything across multiple GPUs (especially the FFTs), switching to one
> process per server with thread per GPU (Volta is the first GPU where how
> one invokes things matters), and atomic synchronization between kernels
> instead of returning to the CPU*."
> 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 2:57 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" -
> > a20171027PD200.html
> >
> > 2) "Nvidia is cracking down on servers powered by Geforce graphics cards"
> > -
> >
> > 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 above articles 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 mailing list
AMBER mailing list
Received on Fri Dec 08 2017 - 12:30:03 PST
Custom Search