Slashdot Mirror
Asus Releases Desktop-Sized Supercomputer
angry tapir writes "Asustek has unveiled its first supercomputer, the desktop computer-sized ESC 1000, which uses Nvidia graphics processors to attain speeds up to 1.1 teraflops. Asus's ESC 1000 comes with a 3.33GHz Intel LGA1366 Xeon W3580 microprocessor designed for servers, along with 960 graphics processing cores from Nvidia inside three Tesla c1060 Computing Processors and one Quadro FX5800."
and it's much cheaper and more effective than just using multiple multi-core processors. parallel computing is the future. how long before we have three dimensional processors?
How many pets would I have to eat to balance out the carbon footprint of this?
I've got a six-pack of kittens at the ready.
Be you Admins? nay, we are but lusers!
manufacturers around the world fear a race to the top!
A note in Asustek's literature said the ESC 1000 has a cost structure in software and hardware of US$14,519 over five years, but an Asustek representative declined to give a per-unit price or when the ESC 1000 would be available globally.
Mmm. 14.5k to make and I'll assume they want to make a profit here. 25k? 30k? Possibly more?
So hey, how much does a regular supercomputer cost? And how do you cool these things?
As a participant in the Milky Way and SETI projects for BOINC, I can say this development is impressive and would be a cruncher's dream come true. It would put supercomputing power in the hands of the everyman and allow applications that rely on distributed computing to take a leap forward.
Very impressive, but you could get something very similar last year.
This grade of machines need Linux on them... not Windows; and Asus has been in bed with MS for some while now.
If you keep throwing chairs, one day you'll break windows....
http://www.youtube.com/watch?v=7Eb1yih5kNY
I remember when that ad came out. I was so pissed. Apple preys on people who have no concept of the scale of computing and this campaign really got under my skin. Now I just laugh at it, but they're still advertising this way, with their comparison charts and graphs touting biggest and best with comparisons to competitors' computing hardware from years past.
No sig for you. YOU GET NO SIG!
The real question of course is, what the "Windows Vista experience index" of this machine is. If it's anywhere below 5.5 it's obviously not worth the bother.
Ummm isn't this just a ridiculously powerful desktop computer rather than a super computer? The current 500th super computer on the top500 list is this machine which has a Rmax of 17 Tflops and an Rpeak of just over 37.6. Now its impressive that this desktop system has 1/37th of the power of the lowest machine on the super computer list... but does that really make it a super computer? Moore's Law says that it will take around 10 years for this desktop box to evolve to the power of that current bottom top500 box. So in other words its 10 years behind the performance of the current 500th best super computer.
If its because it hits 1 Tflops then in a few years time you'll have mobile phone "super computers" as Moore's Law is still moving onwards.
This is a very very fast desktop computer suited to certain simulation elements which are GPU intensive. Nice box, fast box.... but not a real modern super computer.
An Eye for an Eye will make the whole world blind - Gandhi
The Tesla c1060 processor boards sound like a very efficient way of packing in compute power, but unless they're neglecting to mention it, the 4GB of GDDR3 RAM each has on board has no error correction. Given the rates of correctable errors observed e.g. here, I could never recommend using it for computing simulations that matter. A flipped bit in a floating point number can have a disproportionate affect on the outcome of calculations that rely upon it, and short of running the whole simulation a second or third time, one couldn't be confident that such an error did not occur.
Large compute-intensive simulations can take weeks, and are used to justify engineering and business decisions that involve the disposition of large amounts of money and other resources — it is important that the computational part of the process can be relied upon.
While I wouldn't choose to do my scientific computing on Windows, I know some people do, and those Tesla cards (which are providing the bulk of the processing power) really don't care which OS you're running.
[b.belong('us') for b in bases if b.owner() == 'you']
One case of them kittens would be equivalent, or you could substitute one PITA mother-in-law. Would you like some ketchup with your soylent green? It's a nice color contrast.
The Cray T3E-1200E reached 1 teraflops in 1998. Now, we can reach that same level of performance (depending on the app) with a desktop computer. How time flys...
Life is not for the lazy.
how I long for the good old days, where a supercomputer meant a sexy cray, Sgi, thinking machines blinking led monster. Its just not the same when a supercomputer basicly is the same hardware as we have at our desk. Seems anything today can be a supercomputer, even the ugliest box.
mod this up. So truth, seeing problems like this in very "not life saving industry" every day
God's gift to chicks
Is it any wonder that US economic dominion is eroding the way it is. This is yet another telltale sign of the fact that China is going to be the economic superpower of the 21st century not the US.
Sounds like a nice toy to run stuff coded for CUDA or OpenCL - does anything OS than OS X support either of those properly yet?
Village idiot in some extremely smart villages.
The PSU is only 1100W. It's not that intensive - three teslas are like three big graphics cards. 2 or 3 kittens would be sufficient, so you've got enough to share.
1100W? Can I eat my vacuum cleaner instead? Yummy.
Do you have pepper sauce?
Pepper sauce? Pepper sauce?!? Do you have any idea what the carbon footprint of pepper sauce is? My brother ate pepper sauce once. He had to eat a whole zoo full of animals to make up for it! Stay away from the sauce!
These posts express my own personal views, not those of my employer
processors to attain speeds up to 1.1 teraflops.
So you're saying it's fast enough to run Windows 7, but forget Vista?
These posts express my own personal views, not those of my employer
This is :
http://helmer.sfe.se/
Domestic spying is now "Benign Information Gathering"
I put it to you that any computer that fits on or under a desk is not "super".
...which will be used principly for... typing e-mails and surfing the internet, just like 90+% of other desktop computers... oh yeah, and downloading lots and lots of porn. Way to go, guys! Keep the hits coming!
How much is this Swedish(Taiwanese) Made Pen15 enlarger gonna Cost?? It maybe "desktop" sized but I doubt desktop prices!
In response to any question: I'm not sure, let me consult my super computer and get back to you. In any presentation: After crunching a lot of numbers on my super computer I can tell you that ...
And you're saying this...why? Are you somehow convinced that these processors show up as general purpose CPUs? They don't. There is no conceivable reason something like this "needs" Windows. You're going to have specialized compilers generating specialized code that gets handed off to the GPUs. OS is mostly a non-issue.
Finally, I can play Second Life at full framerate....
Doesn't the GTX 295 alone put out 1.8 TFLOPS? It can't possibly be much different in architecture to the cards in this "supercomputer"... Slap in a different BIOS and you'd have the same thing, but significantly cheaper, right?
Before or after she died?
Nah - with such processing power, one might actually see a Windows machine perform properly! From boot to blue screen of death in mere milliseconds! Run your malware faster than ever! See clippy dance furiously across the screen in smooth 250 fps animation!
Wow, just thinking of how quickly I could compile text-based monopoly or fortunes, from source, on Debian here... No, it may not be a true "Super Computer", but it'll sure play a mean console game, won't it?
Holy happy hippy crap!
Imagine a beowulf cluster of these...
Why does it have such an ancient and shitty graphics card? The GeForce FX series were terrible in their day and not worth the gold in their circuits today.
You know if every computer becomes a super computer, there wouldnt be any super computers.
Since when has a workstation processor been considered server hardware?
San somebody who has actually worked with such machines enlighten me about its performance on tasks that are not floating point intensive? Our simulations mainly push many,many objects around, with relatively little, or no floating point math in them.
Do such machines still make sense, or are we better off with a bunch of general purpose CPUs clustered together? How do they compare to Suns Niagara cpus that have umpteen hardware threads in them ?
Complexity is a measure of our ignorance...
nice to have powerful machines. But what about the programming end ?
More specifically, can it run MATLAB or Octave and use all the flops for computations ?
I think its a known fact that most academia use MATLAB/Octave to do model creation/testing...
During the transition.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
This grade of machines need Linux on them... not Windows; and Asus has been in bed with MS for some while now.
it has support for red hat and suse...
Now Microsoft has new minimum desktop specs for it's next iteration of the Windows ...
Does it run windows 7?
n/b
... imagine a Beowulf cluster of these!
But if you run Linux on it, you have to deal with the Nvidia drivers.
There's no failure quite as dissatisfying as a complete and total solution to the wrong problem.
Oooo, thanks Stanley, you hunk of man you, but I said pinch, not punch!
Yes indeed. Who in their right mind would run anything GPU-intensive on Windows? The platform is well known for having absolutely terrible video drivers. I hear that most manufacturers don't even support the platform, and just expect the community to write drivers!
I am TheRaven on Soylent News
Imagine a beowulf cluster of these....
I was expecting it to be called the Eee-1. But EEE-niac would have been cool too.
It seems like loading up a motherboard with loads of PCI Express slots coupled with the 5520 chipset just makes for trouble. I actually almost bought the motherboard upon which this computer is based, P6T-WS-Professional, but the problem is that it got some fairly mixed reviews as far as stability goes. Tyan has a similar product, the S7025, but let yous you use two CPUs. In both cases, people are reporting issues with.the boards.
It's rather unlike ASUS, for sure, as I trust the brand of motherboard. So I stepped down to a less exotic asus z8pe-d12, which, has the added bonus of letting me run dual xeons, rather than just the single. I probably could have gone with the Tyan, but it was more expensive, and honestly, GPU computing isn't something I'm doing.
Incidentally, not having PCI Express lanes like this is probably killing AMD more than anything else right now. I looked at building an Opteron board based on Shanghai / Istanbul instead of Nehalem but the difference was PCI Express x16. There's not much out there at all for AMD that supports it and dual CPU. It's a shame because I had an Opteron previously and I like the brand a lot.
This is my sig.
BSOD/Clippy jokes? Sure is 1998 in here.
From TFA: "Computers able to perform at such high speeds can be used in a variety of ways, including scientific research, image manipulation, engineering modeling or for medical purposes."
Medical purposes? Really? OK, so I vote this is a better solution than a shot in the arm against the pig sniffles. Give them away for free, government sponsored!
It's rather unlike ASUS, for sure, as I trust the brand of motherboard.
I automatically bought ASUS motherboards by choice, and that's usually worked out well, but the last two models I've bought have been nightmares... the M2A/VM and it's predecessor model (I forget the model number now). I bought the older model, and was unable to use but a single channel of RAM. The store "upgraded" me to the M2A/VM, and after replacing it twice... the store was unable to get the first two to work dual channel even with their own CPU and DRAM both times... I was finally able to get it working with the third M2A/VM. For about a month.
Would I buy another ASUS motherboard? Probably... none of the other manufacturers have had a perfect track record either. But I don't expect them to "just work" any more.
What would be the benefit of one of these machines vs a 1920 core, 4x GTX 295 machine, if I do not need the extra memory on the GPU?
Why are you all picking on Windows?
Gnome and KDE are much bigger fat cats when it comes to memory footprints.
... to speed up web access on an HTTPS-only web site?
now we need to go OSS in diesel cars
and have up to four teraflops ... http://www.nvidia.com/object/tesla_supercomputer_wtb.html
now we need to go OSS in diesel cars
The fact is, 1998 was a golden age of computing. It was the time before every GUI had to have gradient colours and shiny marble styling, and stupid 3D effects.
...that the chipset is passively cooled!
Does it run Linux?
Now , not only can you fold the dna markers, you can also hotbox about 960 WoW accounts all at the same time....
groovy, I want one for christmas
Remember that this 1.1TF is single-precision; double-precision is around 240GF. Let's hope they fix this in the next version.
Also, there is 240 cores per C1060, for 720 cores total of Tesla power. The additional 240 cores come from the Quadro in the system; those cores may occasionally be busy with graphics work and unavailable for computation.
It's Linux, damnit! Pay no attention to renaming attempts by self-aggrandizing blowhards.
does anyone have historical comparisions going back to the 70s, eg, how many terflops and how much ram nasa had during apollo.
I have this memory of an ad taken out by Boeing in the late 70s, offering their world class supercomputers to researchers; among the leading edge attributes was 500 meg of solid state memory
While this sort of machine is useful (I just built one for quantum Monte Carlo calculations 6 months ago) it is hardly news. NVIDIA has been pushing this sort of machine since the launch of the Tesla. In fact, they have had a parts list on their website for some time telling exactly what is needed to put together a computer with 4 C1060's. This is not even the first commercial offering of this nature, with companies like appro and microway having similar products for at least a year (see nvidia) for a more complete list.
Nvidia has had a page up for a while on this. Most of these units use small desktop motherboards, have limited ram and IO capability, and lots of GPU. These are poor designs for many calculations. These guys have a dual socket, full server class motherboard with up to 144 GB of ECC DDR3 RAM, as well as putting more than 500 MB/s into their local disk IO channel with up to 32 2.5 inch SATA or SAS drives, in a single quiet deskside chassis.
Asus wasn't the first. They are about a year late to the party.
What's so bad about NVidia's Linux drivers? Yes, yes they may be binary blobs but they work quite well.
There is nothing stopping you from making an Open Source 6 pack of bonsai kittens...
Does it run Linux?
cualquier vaina hagase el muerto
What are you talking about? In 1997 I had windowmaker running on my NetBSD machine with a gradient theme, marble pixmaps for my dock tiles and things exploded when they were pulled from them the dock.
Before that I used AfterStep 1.0.
Your thinking of '88 not '98.
Have you used the Nvidia drivers under linux, CUDA is about the only thing they're good for
Any and all content posted above may be ignored, considered irrelevant, or otherwise dismissed.
For some reason this story is tagged !supercomputer (as well as supercomputer), which seems downright churlish.
Sure NVidia's CUDA architecture is quite specialized and has some severe constaints, but OTOH so do any of these modern cluster-type supercomputers. Certain types of application map well onto these architectures and others don't. The CUDA architecture is certainly more constrained in terms of memory access per node and inter-node connectivity than say a cluster of Linux nodes.
OTOH, look at the downright mind-blowing maximum FLOP rating of this beast - 1,100,000,000,000 floating points ops per second!
Putting that in perspective I remember c.1980 when the DEC VAX was considered a very powerful department level computer given it's 1 (one) MFLOP rating - that's only one millionth of the power of this NVidia beasty!
Of course it remains to be seen what sort of FLOPS anyone can achieve in a real-world application on this, although presumably it would do pretty well on graphics rendering for which the architecture was originally designed.. be interesting to see how it compares on the types of graphics rendering that CRAY supercomputers were previously used for.
960 graphics processing cores So basically, it meets the hardware requirements for Vista, even with Aero enabled!
Buy one of these things and run it 24/7, and you'll never go cold in the winter.
Does it run Windows? I really need my solitare. How fast does it start office? Does it have those cool lights in the fans? These are the important questions we need answered.
Who would win this election: Andrew Weiner vs Andrew Weiner's weiner.
Who modded this guy up??? They work great if you use an OS that is a couple of years behind the curve and don't care about kernel mode settings or freedom. But run a cutting edge distro and their crappy driver only works for the second half of the release cycle. I got so sick of loosing my graphics that I swore off all things NVidia and ATI/AMD gets all my buisness now.
There you go, using anecdotal logic again.
Michael J. Ryan - tracker1.info
That's interesting. I am running Fedora 11 which I would say is on the edge, and I have newer had any problem re-installing the nvidia drivers(take less then a minute) each time the kernel is updated.
Your sarcasm might be true for gaming, but Linux rules on scientific / number crunching applications, and that is what Tesla is built for. From Tesla Tech Specs, Windows XP is not even supported on the higher end S870 model.
with that xorg.conf file.
To be fair though, the drivers do sometimes lag behind the newest kernel release, but this is easily mitigated by reverting back to the last kernel where the drivers were working properly, or by waiting a while before installing new kernels.
Once I was a four stone apology. Now I am two separate gorillas.
and see what the Iixians come out with.
My point was really that the OS is completely irrelevant, as it is on a few IBM 'Linux' supercomputers that run Linux on PowerPC cores that handle job scheduling and I/O offload and do the real work on another processor. This machine will be doing all of its real work on the GPU, not on the CPU. The OS loads programs and data to the CPU and then gets out of their way. You could be running DOS on these machines and it would make no difference to the throughput, only the 3D drivers (which, these days, include a JIT compiler going from platform-specific bytecode to GPU-specific instructions) matter.
I am TheRaven on Soylent News
So, if 1.1 the theoretical max tflops on this thing, and assuming we're talking about double precision operations.. wouldn't 2 radeon hd 5870s match it? In theory, at least.
From HD5870 specs:
# Processing power (single precision): 2.72 TeraFLOPS
# Processing power (double precision): 544 GigaFLOPS
Now, whether or not you can use all that power through GPGPU may be a different story.
If they can call a custom desktop PC a supercomputer, because it has specs that used to be in the range of supercomputers, then my wristwatch is also a supercomputer.
(More pictures, including the obligatory model to hang off the computer like it was a sugar daddy.)
I could get that with two 9800GTX+ cards.
Too much hardware here for too little performance.
Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
I have on my desktop a double socket W5590 with 4 Tesla C1060. You can get one of those systems from E4 in italy. http://www.e4company.com/
I've seen this kind of crap before. "Super speed" they cry out. But programming environment? How about the fun of re-writing all of your apps from scratch? And games? Games won't run on this! Games need a processor, and also a GPU. This has GPU galore, but processor? Nada. And it likely won't run well with Linux. Without Linux, its a non-starter. You can say 'but, but, but..." all you like, but in the world of supercomputers, Linux is where its at (and there are a LOT of reasons for that, price being the last). Asus has earned a really lousy reputation in the Linux community. This dog don't hunt.
Excuse me, BUT...
.385K multiplies (yes that's point 385K) per second. The atomic bomb was pre-ENIAC. I don't think you need a teraflop of computing power to magically get your Bomb.
The first hydrogen atomic bomb was designed back in the days of ENIAC, which could perform
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
after 2 weeks my windows will run slow anyways.
[s]
AA
It doesn't say the precision of the flops. Unless things have recently changed, the outlandish flops spec of GPU derived compute platforms have all been 32-bit precision, whilst the TOP500 score is explicitly 64-bit precision.
http://en.wikipedia.org/wiki/FLOPS
nVidia's Tesla C1060 GPU computing card performs around 933 GFLOPS in single precision calculations
the same Nvidia Tesla C1060 capable of 78 GFLOPS in double precision
XML is like violence. If it doesn't solve the problem, use more.