NVIDIA and AMD Launch New High-End Workstation, Virtualization, and HPC GPUs 95
MojoKid writes "Nvidia is taking the wraps off a new GPU targeted at HPC and as expected, it's a monster. The Nvidia K20, based on the GK110 GPU, weighs in at 7.1B transistors, double the previous gen GK104's 3.54B. The GK110 is capable of pairing double-precision operations with other instructions (Fermi and GK104 couldn't) and the number of registers each thread can access has been quadrupled, from 63 to 255. Threads within a warp are now capable of sharing data. K20 also supports a greater number of atomic operations and brings new features to the table including Dynamic Parallelism. Meanwhile, AMD has announced a new FirePro graphics card at SC12 today, and it's aimed at server workloads and data center deployment. Rumors of a dual-core Radeon 7990 have floated around since before the HD 7000 series debuted, but this is the first time we've seen such a card in the wild. On paper, AMD's new FirePro S10000 is a serious beast. Single and double-precision rates at 5.9 TFLOPS and 1.48 TFLOPS respectively are higher than anything from Intel or Nvidia, as is the card's memory bandwidth. The flip side to these figures, however, is the eye-popping power draw. At 375W, the S10000 needs a pair of eight-pin PSU connectors. The S10000 is aimed at the virtualization market with its dual-GPUs on a single-card offering a good way to improve GPU virtualization density inside a single server."
My entire computer uses less power than one of these cards.
Intel's Xeon Phi is also out (Score:3)
How long is the wait ? (Score:5, Interesting)
Right now they are all too expensive, and consume too much juice.
How long would the wait be before these things get to have pricetag that average Joe (well, advance version of average Joe) can afford ?
Re: (Score:2)
Re: (Score:2)
3 years is right, considering theyre about 3 generations behind AMD and Intel.
Re: (Score:3)
Re: (Score:3)
I'm hoping for a consumer-level version so I can run Linux as my primary and Windows as a secondary gaming OS and still get full GPU performance on both platforms.
Re: (Score:1)
You can already do this. Search for "PCIe passthrough" "VGA passthrough" "Xen passthrough" etc. Requires a processor that supports AMD-V or Intel VT-d.
Desktop Virtualization does graphics on a server (Score:5, Informative)
Server virtualization doesn't really need this - running web servers or databases or name servers, which are all essentially fancy timesharing.
But "Desktop Virtualization" emulates your entire desktop as a virtual machine on a shared server, graphics and all, and just ships the rendered screens back to your desktop, accessible from anywhere, with RDP or VNC or whatever, kind of like a clumsy version of X Windows except you get to do full-scale graphics acceleration at the server farm instead of at your desktop. The mainframe IT crowd like it, because the PC on your desk can be dumb and low-powered, and the server back in the server farm they get to maintain can be big and fancy, and they can have better control over it than over your desktop, don't need to keep every bit of software up to date on everybody's remote PC, and it's generally easier to manage. And if you're logging into your work desktop from Starbucks, they don't have to protect it as thoroughly from everybody else there, and you can access your work Windows desktop from your personal iPad or your kid's gaming machine or whatever, and the company data's not very vulnerable because there's really nothing running on the remote machine.
And using this chip, they've got a lot more graphics horsepower available for rendering desktops, so for instance they can provide you with adequate performance for video editing, not just for email and word processing .
Re: (Score:2, Interesting)
I think you missed the point here. Even if the card is capable of doing super high res rendering, the network traffic would be the weak point. The graphics card is much much better off being put in the client than rendering to a bitmap that has to be transferred. (And remote desktop protocols try to compare what has actually changed in order to send the least amount of data, which will cause high processing load when you run something at a high FPS).
I think is for using GPUs for processing (Password crac
Re: (Score:2)
I didn't say I'm convinced that Desktop Virtualization is a good thing, as opposed to a scam for selling hardware to IT departments. But this chip could very well make it better. It seems to be especially popular in environments where everybody has an actual desk they sit in front of, but it's also starting to address the BYOD market.
Re: (Score:1)
The problem almost always has everything to do with the technologies that exist to send screen data to the zero/thin/thick client over anything less than a sub-5ms, 1Gbps+ LAN. RDP is trash, VNC is trash, PCoIP is slightly less poopy trash.
Now, could this GPU have some sort of application for, say, the video editing industry, where these client/server systems are
Re: (Score:3)
"Can use" =! "works well".
Re: (Score:2)
My standard desktop environment at work uses RDP on a 10 Mbps LAN through an old slow firewall to connect to a jump server in my lab; when I'm at home I use SSL VPN over 3 Mbps DSL to get to the company network gateway location, and then run RDP over that through the firewall in my lab. (We do computer security development in our underground laboratory, so we can't simply connect it to the office LAN because it has a direct internet connection with various mean nasty ugly stuff on it.) The jump server use
Re: (Score:2)
Use the GPU for heavy compression of bitmap data? :)
Re: (Score:2, Informative)
I've done several Virtual Desktop projects for "scientific users" using GPU based acceleration on the physical server hosting the VDIs. The benefits of a powerful GPU are many:
1) You can provide GPU accelerated applications to all or some VDI users (I'm talking about GPU for calculation, not necessarily rendering graphics).
2) You can provide accelerated rendering by pairing server-GPU with whatever GPU is on the endpoint desktop for a better user experience (especially over limited network bandwidth), using
Re: (Score:2)
1) The orginal poster had said "except for GPGPU etc."; there's lots of obviously cool computational application for these things, if you need to do heavy calculation.
2) Thanks, I hadn't known about that! (I haven't really looked at Xen since Citrix bought it, though I'm starting to see people using it instead of VMware because of the Free Beer prices.)
3) This thing seems to be about 5-10 times the price of a typical medium-high-end video card, so you need a reasonably large group of users to make it pay
Re: (Score:1)
Its too bad that GPU acceleration on the VM world is so infantile right now. Hyper-V and VMWare can sorta do it OK but again, it has its issues and problems such as overhead or just not running all together (certain apps want certain cards.)
No one in a business is going to drop $4000 on a card to render kitty videos faster for the accounting department, no, these are for CAD and design engineers who want to go thin client.
My entire house uses less power than one of these (Score:1)
Well, it's good to see AMD keeping with tradition, a blessing for those in the frozen north..
Re: (Score:1)
Voltage isn't power..
Re: (Score:2, Informative)
so 375 watts at 12 volts is the same as 375 watts at 120 volts?
Yes, unless you're talking 375 amps at 12 volts vs 375 amps at 120 volts (which is quite a lot)
Re: (Score:2)
Ummm, but he isn't, he's talking watts...
Re: (Score:1)
Re: (Score:2)
Yes, unless you're talking 375 amps at 12 volts vs 375 amps at 120 volts (which is quite a lot)
Did somebody seriously mod that "informative"??? Sheesh.
(Right now it's modded "+2 infromative"...)
Re: (Score:2, Informative)
Ugh...
Repeat after me, Watt/hour = Amps/hours * Voltage
Amperage = Current over time
Wattage = Power over time
Current(Amp) != Power(Watt)
Btw, if you don't know, normal electrical devices are measured in hours, meaning X watt usage really means X watts over a period of an hour.
So 375 watts = total power
Higher voltage and lower current can equal the same power as lower current and higher voltage. Realistically, certain applications require certain voltages to work (high voltages for ovens as they are basicall
Re: (Score:2)
No.
Amperage is synonymous with current, wattage is synonymous with power...not current over time or power over time, that would be amp-hours or watt-hours if you're bringing time into the equation.
Correct.
No.
Batteries and other electrical sources have capacities measured in
Re: (Score:2)
Have to correct myself, too...
Re:My entire house uses less power than one of the (Score:4, Informative)
so 375 watts at 12 volts is the same as 375 watts at 120 volts?
Yes, unless you're talking 375 amps at 12 volts vs 375 amps at 120 volts (which is quite a lot)
I think it's better to say 31 amps at 12 volts and 3 amps at 120 volts, both totaling to 375 watts.
Re: (Score:1)
I think it's better to say 31 amps at 12 volts and 3 amps at 120 volts, both totaling to 375 watts.
Thank you for saying this, I think my head was about to explode from the people trying to figure out how it stuff works.
Re: (Score:1)
Yep, 375 watts is 375 watts... Pretty amazing, isn't it?
Re: (Score:1)
"Watts" are a unit of power, which is a measure of "energy" delivered per time. The technical notion of "energy" is actually fairly different from what a non-physicist probably thinks of when they say that word. "Voltage" or electric potential is closely related to energy, and is measure by the unit "volts." It is, however, not an energy. 375 Watts is a certain amount of energy delivered every second, in particular, 375 Joules. If this provided by an electrical current at 120 volts, or 12 volts, this is sti
Re: (Score:2, Informative)
Volts = Difference in height between two water tanks
Amps = Flowrate in the hose connecting both tanks
Watts = Number of full water buckets * height difference you'd have to carry _per second_ to do the same
Energy = Total volume of water transfered * height difference
Re: (Score:1)
Re:My entire house uses less power than one of the (Score:5, Interesting)
just wait until you wet your noodle on capacitive/inductive AC where watts is watts except when it's volt-amps, and efficiency is measured as a ratio between watts and volt-amps.
Re: (Score:2)
unreal!
Re:My entire house uses less power than one of the (Score:4, Informative)
and efficiency is measured as a ratio between watts and volt-amps.
No, efficiency is still measured as useful work out / Joule in.
You are referring to power factor, not efficiency.
The two are related, since electricity companies bill you on the number of Volt-Amps * Time used, not number of Joules, since it's much harder to do the latter, and the former determines the actual current which determines most of the expensive things, i.e. transmission losses, wire guage (infrastructure costs), etc.
If you're seriously screwey on your power factor, you can buy a power factor corrector and it will probably pay off quite quickly if you're a big user.
Re: (Score:2)
Power factor is a useless measure anyway. It is just a ratio. A bad power factor at 2W is not as bad as a bad power factor at 200kW. Use Vars instead. A much more useful value to have. Given the math of it I fail to understand why anyone is interested in PF. Source; I have designed PF correction systems (up to about 10MVars).
Re: (Score:2)
WRONG. Electric companies bill on watt-hours or other equivalent units of energy. Although large users of electricity may be charged penalties if the power factor is bad (V*A>>W), the basis is energy, not volt-amps*time.
Re: (Score:1)
In terms of power, it is.
The reason why this is a bad idea is that the currents you'd need then are extremely high (375 W at 12 V is 31.25 A, 375 W at 120 V is 3.125 A), and high currents means high losses in the electric line delivering the power (the losses grow quadratically with the current), which means you'd need thicker wires to compensate (the resistance is inversely proport
Re: (Score:2)
the resistance is inversely proportional to the cross section)
It's worse than that. At high currents, you need large cross sections. With AC, the skin effect will start to kick in eventually, which makes the resistance inversely proportional to the circumference of the wire, i.e. with the square root of the cross section.
Which is really annoying.
Re: (Score:2)
Would it be an Earth worth living in? What's the point of saving it if there is nothing to do.
Quick! (Score:1)
Re: (Score:3)
That's so last century.
These days it's, "imagine a Bitcoin mining rig of these"
Doesn't quite have the same cadence, though,
And they have an 8-board FirePro system running (Score:5, Interesting)
3.6 kilowatts, 16 GPUs:
http://fireuser.com/blog/8_amd_firepro_s10000s_16_gpus_achieve_8_tflops_real_world_double_precision_/ [fireuser.com]
and to think all this comes from video games.
Re: (Score:1)
A thing of beauty indeed.
If you look carefully at the third picture you'll see that the graphics cards are hooked up to a green pcb like thing, and if you look at the first picture, you'll see that this green thing is plugged into the powers supply on the left. And in the second picture you see 3 power cords plugged into the power supply. I dunno where you buy this stuff (or if AMD built it themselves), but I really want one.
Re: (Score:2)
Looks like 3 sepperate PSUs, don't think thats too hard to come by.
Re: (Score:1)
Re: (Score:1)
Re: (Score:2)
"and to think all this comes from video games."
The word "video games" just hides the fact that videogames are SIMULATIONS/Models (although simplified) of some aspect of the world. No one would be surprised since videogames are basically alternative world simulations and we're heading towards a time (eventually over the long term) where extremely complex behavior will be simulated.
Re:And they have an 8-board FirePro system running (Score:4, Interesting)
Those 8 TFLOPS would have landed it somewhere at the top of the #500 supercomputer performance list in November, 2011 [top500.org]. ASCI White [top500.org] used 8192 375MHz Power3 cores to achieve this performance. It took up a fair bit of space [energy.gov] and used 3 MW to run the machine with a further 3 MW needed for cooling. It had a theoretical processing speed of 12.3 teraflops.
Re:And they have an 8-board FirePro system running (Score:5, Informative)
Of course that should read 'November 2001', not 'November 2011'...
Ya well there are some limits (Score:2)
GPUs are powerful because they are more limited than CPUs. They are very good at what they do, but not as good at general operations. So, you can find things they are exceedingly good at, and thus way faster than CPUs, but also things they are bad at, and thus way slower.
1 TFlop on a video card isn't the same as a TFlop on a CPU in terms of the things you can do. A simple example of the limits are memory, the GPU relies on very fast local memory to do its work, but it is small, relatively speaking, under 10
Re: (Score:1)
Re: (Score:3)
SHARING IS CARING!
375 W (Score:5, Insightful)
Does the person who wrote this know how much a TFLOP actually is, let alone 5.9 TFLOPS (single precision) and 1.48 TFLOPS (double)? As an example, an Intel Core i7 980 XE does 109 GFLOPS double-precision. This is over 13 times that! It is really exciting to see the power of GPUs broadened to scientific computing in general. I doubt these cards would be cost-effective or are really intended for gaming.
Re: (Score:2)
Re: (Score:2)
Ironically, I don't think you can. At least, not at all easily.
The FirePro cards shows don't support multiple GPUs driving a single output. So you'd have to find some way to avoid treating the array of screens as a single screen, which is far easier said than done. At the very least, extensive reworking of the game's renderer seems in order, and any game open-source enough for a modder to do that is old or graphically-limited enough that you could drive quad QHD monitors using my laptop.
The Tesla and Xeon P
Re: (Score:2)
Well, that one screen is at 2560x1600. Not that many people have 30" screens. (Then again, Apple's "retina" display may start a long-overdue megapixel war in PC displays, if we are lucky).
Also this is "VHQ" quality. I see nothing wrong with releasing a new top title that cannot be played smoothly at max settings with current cards, so long as the game looks at least as good as any oth
There is no "realistic HPC workload" (Score:3)
Re: (Score:3)
Does the person who wrote this know how much a TFLOP actually is, let alone 5.9 TFLOPS (single precision) and 1.48 TFLOPS (double)? As an example, an Intel Core i7 980 XE does 109 GFLOPS double-precision. This is over 13 times that!
By way of comparison, the Opteron 6174 can hit a bit over 180GFlops in LINPACK. On a shared memory machine multi socket machine, the efficiency is very high, so a quad socket Opteron 6100 box would probably manage in the region of 700GFlops.
Anyway, plugging the numbers, the 6174
Re: (Score:1)
Does the person even know that computer components are not sitting there consuming their tdp all the time?
The tdp is just a theoretical number that represents the maximum amount of heat energy the package is capable of dissipating without being damaged. It is NOT the amount of power the device will consume in normal operation. In practice, you will almost NEVER see a CPU or GPU consuming its tdp.
Re: (Score:2)
I don't know hey. Given that these things dynamically overclock parts to when there is thermal/power headroom available, I would guess, yes, you would see it using tdp. Most processors under the right load should see tdp or close to it in the right loads. It might be tdp minus 2% or 3%, but close enough. If you have better cooling or a lower ambient, you might run into the power limit, which is generally set close to the rated tdp. But it all depends what your load is.
Re: (Score:3)
This. Remember the Cray-1 supercomputer? $5-10 million, 100KW power draw. The standard against which other systems were compared through the early 80's.
This card provides more than ten thousand times the computational power, using less than one-half percent the electrical power.
The future really is a cool place.
Nice! (Score:2)
With the adoption of GPU-based rendering in 3D graphics workstations for the entertainment industry it's great to see developments like that.
Rats. (Score:1)
Re: (Score:2)
Your GPU card does more than just output video. The GPUs on these cards are designed for brute force calculations and chugging through numbers. They're designed for physics engines, running through protein folding calculations, and rendering high quality video in real time.
If you want to equate them to a design in the past, think of them as really really really powerful math co-processors. CPUs are designed for short command queues and calculations that are hard to predict the next step, while GPUs are d
Export restrictions (Score:3)
These things are regulated. Argentina tried to buy a few (5!) of their previous line, I think it was Tesla. The US government wouldn't allow it. Guess you have to be a NATO member to buy these.