<br><br><div class="gmail_quote">On Mon, Dec 17, 2012 at 8:15 PM, Lux, Jim (337C) <span dir="ltr"><<a href="mailto:james.p.lux@jpl.nasa.gov" target="_blank">james.p.lux@jpl.nasa.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I wasn't thinking so much about code efficiency, more "wall plug power" efficiency. The board may consume 250W, but it will take non-zero power to support that board, and then the power supply efficiency needs to be taken into account. But I suspect the 1 GFLOP/W was more just an "old" "rounded off" number.<br>
<br>
Yes... it's very hard work to get to a real 75 GFLOP/Watt, but that is what DARPA is all about... High Risk, High Reward. Somehow, though, I can't see building a new fab with smaller feature sizes for the paltry sum of 20M. More like they'll do some architecture studies, a pile o'modeling (if we DID invest $1B in a new fab, here's what you might be able to do), and do a bunch of work on things like failure tolerant architectures (if you have a sea of processors, and X% are dead at any given time, how do you write software to run on that sea)<br>
<br>
I wonder what Nvidia chips are used in Audis and BMWs? The video display, perhaps: there's a nifty 3D rendered view of the GPS mapping info in the new BMWs? I don't see a real need for that kind of horsepower in an Engine Control Unit. Maybe in a smart cruise control that does station keeping, or in a collision avoidance system. Actually, I don't really see Nvidia being in the "safety critical" space at all.<br>
<span class="HOEnZb"><font color="#888888"><br></font></span></blockquote><div>I really wouldnt be suprised if its the tegra's that would be in those vehicles. Only thing i coudl think of it would be used in in car GPS or the fancy radio setup with all those features they have, who knows. </div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="HOEnZb"><font color="#888888">
<br>
<br>
<br>
Jim Lux<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
<br>
-----Original Message-----<br>
From: <a href="mailto:beowulf-bounces@beowulf.org">beowulf-bounces@beowulf.org</a> [mailto:<a href="mailto:beowulf-bounces@beowulf.org">beowulf-bounces@beowulf.org</a>] On Behalf Of Vincent Diepeveen<br>
Sent: Monday, December 17, 2012 10:02 AM<br>
To: Lux, Jim (337C)<br>
Cc: <a href="mailto:Beowulf@beowulf.org">Beowulf@beowulf.org</a><br>
Subject: Re: [Beowulf] DARPA issues 20 MUSD grant to nVidia to go from 1 GFLOPS/Watt to 75 GFLOPS/Watt<br>
<br>
On Dec 17, 2012, at 6:27 PM, Lux, Jim (337C) wrote:<br>
<br>
> That could be a notional 1 GFLOP/Watt in a fielded system.<br>
<br>
Even linpack is 70% - 80% efficient on this so should get out oh let's use a conservative 4.5 flops/watt effectively at codes.<br>
<br>
<br>
<br>
<br>
<br>
Note that (to my big surprise) it seems to be the case that the gpu's are effectively getting higher efficiency than Xeon Phi here.<br>
<br>
> The original documents for PERFECT are probably a year or two old by<br>
> now.. but what DARPA is looking for is a nearly 2 order of magnitude<br>
> improvement... Whether they started at 1 or 1.4 or 6 really doesn't<br>
> make much difference to what they're looking for.<br>
><br>
<br>
Yeah well that 2 orders of a magnitude is just 1 order of a magnitude if we start at 6.<br>
<br>
6 ==> 75 = factor 12<br>
<br>
They speak about 7 nm technology in the accompanying document. That's a very conservative estimate, obviously in theory even with todays 2 dimensional way of building (not to mention when things really get 3d), we speak of a difference in theory of:<br>
<br>
(28 / 7) ^ 2 = 4^2 = 16<br>
<br>
Given enough time, engineers will get that factor 16 easily out of transition over the years from 28/32 nm to 7 nm.<br>
Note that 7nm is still far beyond the horizon.<br>
<br>
However if they would have needed to improve current design factor 75 moving from 28/32 nm they use today to 7 nm, that would be a complicated bet.<br>
<br>
> In any case, it's a long way from a manufacturer's cut sheet to a<br>
> system installed in a tank bouncing through the woods..<br>
><br>
><br>
> Jim Lux<br>
><br>
> -----Original Message-----<br>
> From: <a href="mailto:beowulf-bounces@beowulf.org">beowulf-bounces@beowulf.org</a> [mailto:<a href="mailto:beowulf-">beowulf-</a><br>
> <a href="mailto:bounces@beowulf.org">bounces@beowulf.org</a>] On Behalf Of Vincent Diepeveen<br>
> Sent: Monday, December 17, 2012 5:50 AM<br>
> To: Eugen Leitl<br>
> Cc: <a href="mailto:Beowulf@beowulf.org">Beowulf@beowulf.org</a>; <a href="mailto:info@postbiota.org">info@postbiota.org</a><br>
> Subject: Re: [Beowulf] DARPA issues 20 MUSD grant to nVidia to go from<br>
> 1 GFLOPS/Watt to 75 GFLOPS/Watt<br>
><br>
> "todays 1 gflop/watt" ?<br>
><br>
> The K20X delivers 1.4 Tflop nearly.<br>
> If i google it's 235 watt TDP.<br>
><br>
> 1.4 Tflop / 235 = 6 gflops/watt<br>
><br>
> On Dec 17, 2012, at 2:21 PM, Eugen Leitl wrote:<br>
><br>
>><br>
>> <a href="http://www.networkworld.com/community/blog/darpa-awards-20m-nvidia-" target="_blank">http://www.networkworld.com/community/blog/darpa-awards-20m-nvidia-</a><br>
>> stretch-achilles-heel-advanced-computing-power<br>
>><br>
>><br>
><br>
<br>
_______________________________________________<br>
Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org">Beowulf@beowulf.org</a> sponsored by Penguin Computing To change your subscription (digest mode or unsubscribe) visit <a href="http://www.beowulf.org/mailman/listinfo/beowulf" target="_blank">http://www.beowulf.org/mailman/listinfo/beowulf</a><br>
_______________________________________________<br>
Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org">Beowulf@beowulf.org</a> sponsored by Penguin Computing<br>
To change your subscription (digest mode or unsubscribe) visit <a href="http://www.beowulf.org/mailman/listinfo/beowulf" target="_blank">http://www.beowulf.org/mailman/listinfo/beowulf</a><br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br>Jonathan Aquilina