Re:The Nvidia DGX-1 Deep Learning Supercomputer in a Box
by aconz2 on Tuesday April 05, 2016 @05:41PM
(#11434754)
To be fair, they are comparing it to a dual-socket CPU; which is twice as fair as comparing to a single!!
What I was getting more at was: I want to know the relative performance compared to another 8 Tesla box. I know comparing apples isn't good marketing, but c'mon.
Re:The Nvidia DGX-1 Deep Learning Supercomputer in a Box
by cowsandmilk on Tuesday April 05, 2016 @06:39PM
(#11435155)
They gave a pseudo-comparison to other GPUs in the keynote
Re:The Nvidia DGX-1 Deep Learning Supercomputer in a Box
by astrodust on Tuesday April 05, 2016 @07:58PM
(#11435588)
$129K buys you a lot of dual 22-core servers.
Re:The Nvidia DGX-1 Deep Learning Supercomputer in a Box
by pinewurst on Tuesday April 05, 2016 @09:30PM
(#11435961)
What kind of server pricing are you getting? Base servers are cheap, but add high-end Xeons and memory, not to ignore interconnect and I get something like 7 ok configured 1U servers for $129K (2 20 core w lots of RAM, 10GbE NICs and mirrored boot/swap). No interconnect switching. That's for 20 core Haswell because I don't yet have discount pricing for Broadwell Xeons. I'm sure one could do better at hyperscaler discount but this is startup low-ish quantity.
(Of course, a better metric is that it's getting ~56x the performance at probably ~10x the TDP, but that's not surprising for a GPU with the current state of deep learning code.)
To their credit, the thermal and power engineering needed to get that dense a compute deployment is challenging. (bt, dt, have the corpses of power supplies to show for it.) But the price means that it's going to be limited to hyper-dense HPC deployments by companies that don't have the resources to engineer their own for substantially less money, such as Facebook's Big Sur design: https://code.facebook.com/posts/1687861518126048/facebook-to... . And, of course, the academics and hobbyists will continue to use consumer GPUs , which give much better performance/$ but aren't nearly as HPC-friendly.