Nvidia shakes up supercomputer economics
NVIDIA AND MICROSOFT have started pitching the former's Tesla as the überkit for high performance computing by giving developers a hand on the Windows HPC Server 2008 platform.
According to the new bedfellows, their plan is to encourage developers in adopting Windows HPC Server 2008 as their choice of development platform for parallel computing tasks using Nvidia's GT200-based (eventually Fermi too) Tesla systems, while Nvidia provides the developer community with the necessary software in Nexus. Nexus allows developers to work on massively parallel applications in a familiar Visual Studio-based development suite.
The partnership is still being fleshed out, but if we follow Jen-Hsun's lead on Day 1 of Nvidia's Nvision 2.0, er, GPU Technology Conference, you can see what it's building up to.
Nvidia has retained a solid reputation in the workstation and server business due to Quadro and the first Tesla "supercomputer" chips, however traditionally this kind of presence is accessible only to a handful of companies with very, very deep pockets.
Nvidia promises to address the main concerns with supercomputing installations - power consumption and performance - as scientific compute complexes and render farms sucking more power than small cities. This power and cooling problem can be resolved by plugging in Tesla instead of CPU clusters, it claims.
Nvidia spokesperson Kerry Tescher told us, "Tesla fits extremely well in enterprises' energy-saving attitude. GPU computing delivers 10x the performance in the same power budget."
She added, "A desktop system with Tesla GPUs delivers the same performance as clusters with 16 to 32 CPU servers. Similarly, a small Tesla GPU cluster can outperform clusters that are almost 10-20 times bigger and thus significantly save energy."
According to Nvidia slideware, 32 Tesla S1070 chips will draw 27 times less power than their CPU server farm counterpart, in an oil and gas seismic processing scenario. Even a 2.7 times improvement would be a breakthrough, so if the numbers are anywhere near reality it seems Nvidia might be onto a cash-cow here.
Ah, but where does Fermi fit in all this, you ask? Well, although Fermi has just been announced, the Tesla S1070 and C1060 GT200-based GPUs are still driving Nvidia's HPC strategy. Tesla will make use of Fermi down the line, but right now all the numbers handed out are based upon the current GT200 architecture.
Fermi's increased programmability and GPU-as-CPU DNA should make things a bit easier for developers, especially the C++ crowd. While that hasn't happened, Microsoft Research and Nvidia are busy deploying a "large Tesla GPU computing cluster", although Nvidia seems to be only holding Microsoft's hand in this, as it provides no numbers on computational power or the profile of the installation.
Cray, Dell, HP and Lenovo are lining up for some orders, according to sources. µ
Share this:
ShareBusiness Intelligence Software
Customer Relationship Management (CRM)
Enterprise Accounting Software
Enterprise Resource Planning (ERP)
Enterprise Systems Management
NVidia hasn't made a reliable video chipset since the days of the 5xxx hairdryer series many years ago. Why on earth would any business take such chances on this company under the leadership of Jen Hsun? They are telling businesses to run clusters of their cards now as an alternative to tried and true server CPUs. Do they think we are crazy and stupid?
NVidia: prove you can build a video chipset that won't overheat and die within a year and prove the days of covering up your shoddy hardware mistakes are over. Then we can talk.
Fermi calculations are often not accurate.
Astrochicken & Schrödinger's cat
cogito, ergo suum reductio ad ridiculum, tamen quisnam est duco?
Shticks that build more shticks that build more shticks...
We all gonna fries?
Not an Nvidia fanboi (I alternate as they leapfrog), I thought I'd mention my 8800GTS bought when it came out years ago is still purring happily. Factory OC'ed, driving my 37" @1080p, gaming almost every night.
I never had a failure from either of these guys, you may want to check proper assembly procedures.
Btw, not sure why they would want to talk to you.
bob:
that's jolly nice for you. glad your 8800gts is still purring; my last three nvidia gpus all choked and spluttered their way to the great rig in the sky.
so well done for defending a company that's done me out of two laptop's and one desktop gpu due to defective die. i would, in a keeganesque manner, love nvidia to die (like mine), but that doesn't look like it's going to happen. i can only hope they start losing money like i did (and if you think this ever so slightly unhinged rant is, well, unhinged, then you ain't seen nothing yet. i'm chock full of hate and bile. shit, you would be with this haircut (thanks bill).
ruthlessness in business is one thing, but ripping people off that can't afford it is tantamount to theft.
by the way bob, what the flying duck is suggesting that nvidia's faulty products are due to their consumers '(im)proper assembly procedure(s)' all about?
It's amazing that when nvidia is mentioned on forums how all these angry nitwits start screaming that their nvidia card exploded after two weeks, blah, blah blah. My EVGA cards have been fine and my friends as well. Since Nvidia still outsells the competition almost two to one, dominates the professional market and has the awesome Fermi chip coming soon, methinks they protest a bit too much.
Bump gate, you know when NV had a bad manufacturing issue that pissed off a lot of OEMs and anyone who got stuck with one of the bad cards.
It wasn't so much they had an issue, but that they knew it, denied it, and kept selling them to everyone. That's where they lost trust from investors that pay attention. I have an NV card right now, but my next card (Christmas) will be ATI just because of NV's BS business practices.
where is the story on how the fermi card they showed recently was nothing but a very badly executed mock-up?
it should be done by now...
Nvidia Failed this subject so well it is used in the current syllabus.
I agree with Vinster.
Will they give a 3 year "replacement no quibble" warranty for their Tesla cards?
Thats what a server farm manager wuould want if he was going to throww 3 of these in a 2S box and put racks of them in room.
You would want some decent cooling per box or the cards would go critical.
Especially when the current prototype GT300 card has the vents blanked off ... lol !!
Agreed. NVIDIA doesn't play clean when they're losing. The Fermi launch will be a paper launch, and all the hype is to reduce ATI's DX11 monopoly until the hard launch, that is likely somewhere in 1Q-2Q 2010. A deserved score for ATI.
I wish Charlie was here to give his input on it for us INQ followers. Oh well, off I go to visit S|A. :)
Not quite sure I'd put too much of my own "personal supercomputer fund" in this yet. Over the last few months I and four colleagues wound up with shiny new Dell laptops courtesy of Nvidia graphics cards failing on $3.5-4.5K M90 laptops still under warranty.
Nvidia, to my knowledge, has never acknowledged fault. Not a good sign.
So at least for me, it's wait and see time. Or maybe a baby step with dual quad-core Xeons and lots of RAM.