96 Processors Under Your Desktop
Roland Piquepaille writes "A small Santa Clara-based company, Orion Multisystems, today unveils a new concept in computing, 'cluster workstations.' In October, you'll be able to choose between a 12-processor unit for less than $10,000 or a 96-processor system for less than $100,000. These new systems are powered by Efficeon processor from Transmeta and are running Fedora Linux version 2.6.6. Apparently, this new company has friends in the industry. You already can read articles in CNET News.com ("A renaissance for the workstation?"), the New York Times ("A PC That Packs Real Power, and All Just for Me," free registration, permanent link) and the Wall Street Journal ("Orion Sees Gold in Moribund Workstations," paid registration). The company is targeting engineers, life scientists and movie animators. It's too early to know if the company can be successful, but I would certainly have to get one of these systems under my desk. In this overview, I've picked the essential details from the three stories mentioned above."
You could use these systems as such servers. The idea, though, is that these might be cheap enough to allocate to individuals.
No video card. These are just render/compute clusters in a box.
I'm impressed at the claimed 220W peak power consumption of the 12-node box, but wonder what kind of real computing performance it provides.
-Isaac
I am not a lawyer, and this is not legal advice. For Entertainment Purposes Only.
More frequent yes. But there are more parts within one board, so each of them separately needs less bandwidth than all of them taken together. So, 1G carefully engineered/switched (so each part has 1G bandwidth, not 1G shared between all) is quite sufficient. But then, say, 100 parts need 1G bandwidth between each other and 100M bandwidth to the other board, each. Makes 10G of throughput between boards easily.
Anagram("United States of America") == "Dine out, taste a Mac, fries"
the transmeta chips used are specially developed to run at a low temperature. It is in fact this development alone which has enabled these "mini clusters" to now be manufactured
You should probably check out the product description anyway though; there are some quite interesting hardware design decisions in there!
UNIX? They're not even circumcised! Savages!
Warning ! Warning ! Warning ! Warning ! Warning !
Attention, a public service announcement follows : do not read the "overview" touted by Mr. Piquepaille. This person constantly spams Slashdot, trying to get traffic to his site on Radio Userland (which I'm not linking to, for obvious reasons). Do NOT go to his overview, you're only giving traffic to a spammer. See these recurring complaints, for instance. Not to mention he steals the images he puts on his blog and sometimes also spews bullshit for lack of knowing better. This must stop. In any way, do not fall for the spam, and do not provide him any more traffic. Please also warn fellow readers when you see one of his self-serving posts.
And now, a personal message (warning : verbal abuse in foreign language follows) : Roland, tu nous les brises. Va te pendre, hé Ducon !
[disclaimer : I'm not commenting on whether the subject is interesting, or not. But the kind of astroturfing the submitter engages in regularly is just wrong]
Warning ! Warning ! Warning ! Warning ! Warning !
Xenu brings order!
Oracle RAC is one, but I can't think of any other popular title that would, expecially not for the desktop
Apache. It's on this computer I'm typing on now. Over 300 threads. I THINK it might get just a bit faster.
It's called XD1. Used to be manufactured by Octiga Bay, but that company has been absorbed into Cray.
The 12 node system has a peak power consumption of 220 watts. Not a lot of heat, in the scheme of things.
You can get an 8 proc Opteron cluster for about $10k from Rocketcalc. I can't speak for their new systems, but I am currently running some simulations on one of their older 8 node P3 1Ghz clusters. It absolutely blows my desktop P4 3 Ghz out of the water (even with some rather poor parallelism in my code.)
I wouldn't worry too much about it. Even a single processor still tends to outclass the remaining parts in your computer, performance-wise. If you got a machine with 96 processors, you probably wouldn't notice much of a performance difference. i.e. This is only helpful to people who run heavily threaded, CPU intensive applications. Examples include:
- CAD Modeling and testing of areodynamic vehicles
- Modeling of Oil Wells
- Searching for Extraterrestrial life
- Solving very complex math issues
- Running realistic simulations (e.g. explosion modeling)
Sound familiar? That would be Sun and SGI's workstation market.
Javascript + Nintendo DSi = DSiCade
BLAST doesnt take that long anymore.. well at least not for some things.
We use it all the time to compare our DNA products to all known Gnomes. It takes like 30 seconds.
(300 BP search against the whole library takes less than 40 seconds. using http://www.ncbi.nlm.nih.gov/BLAST/)
Maybe its much longer for other things ?
> The search you linked to pulled up a pile of cool articles.
Please read the paragraph where I said I wasn't commenting on whether the article is interesting. The "coolness" of Piquepaille's spam isn't the subject.
> What "spam" are you talking about in this off-topic, flamebait troll of yours?
Please read the part where it says that the "overview", "more details" and so on in Piquepaille's submissions are always directed at his own weblog in an obvious attempt to get more traffic and hence more money from the ads that show on the aforementioned weblog. Posting disguised self-advertisements on Slashdot seems spammy enough to me, but you're of course entitled to your own opinion.
> What's self serving about any of those stories?
The fact that he always uses his submissions to drive traffic to his own site, where he uses plagiarized and/or stolen content to gain ads revenue. Note that all his submissions to Slashdot include a link to his own site. You won't find any where he would just point to the people who wrote the original content without peddling his blog. You also won't likely find anybody except himself posting links to his blog. Can you see a pattern here ? I sure can !
> Do you have anything interesting to say about this interesting looking workstation?
No, and I don't give a damn. See the aforementioned paragraph.
That's it. Posting deceptive self-advertisements to Slashdot should not be rewarded by ad traffic. But if you want to support him, you're of course free to do so. At least, you'll have been warned.
Xenu brings order!
Ummmm no. The spec says 1500W for the 96 processor one, not 150W. The PSU won't be too big really. I have a deskside SGI Challenge L. The PSU is rated at 2500W (!!!) and it's only about 2-3 times the size of a normal ATX PSU.
Like tinyurl, but one letter less! http://qurl.co.uk/
While it's certainly cheaper to pickup 20 Dell PC's for $500 each, an integrated 12-way workstation may signify the beginning of a new desktop computing standard.
When the IBM AT first came out, $10k was the ballpark for what was a single processor at a few mhz. Now we have a dozen procs running at a few ghz in a federated workstation environment.
The application of this should not be understated. While SETI might seem like fools gold, the proliferation of this kind of computing horsepower could dramatically decrease the time needed to find life. Realistic simulations (eg: explosions) require a rack of specialized equipment that this platform obsoletes, imagine what a LANL scientist could do with this.
Just as the Sparc was the impetus for the modern PC "workstation", a 12-way personal supercomputer would open the door for a new level of desktop performance. Imagine a 16-way Dell "personal" workstation for $5k -- it's not impossible to imagine.
Eric Sarjeant
eric[@]sarjeant.com
Here's a 12-node cluster made of VIA EPIA V8000's with 800MHz C3 processors. Idle power consumption is ~140W.
"The machine runs FreeBSD 4.8, and MPICH 1.2.5.2. After working with his machine and running some basic tests, Glen's cluster looks to be equivalent to at least 4 (maybe 6) 2.4Ghz Pentium IV boxes in parallel on a similar network - achieving a performance of around 3.6 GFLP."
In the end, why put this thing under your desk? Just leave it in the server room and enjoy the quiet.
The arrangements of the pins are slightly different, preventing punters from simply snipping off a pin, and using their old Athlon64 CPU in a shiny new motherboard. As I understand, the Athlon64 chips integrate a memory controller onto the die, so every time the memory controller design is changed, the socket must also change.
The 940 pin Athlon64 required expensive, slower ECC ram, just like the Opteron. But unlike the Opteron, the Athlon64 was intended to be sold to people who might not appreciate the tradeoffs involved in using ECC. AMD corrected this oversight, and then released a new socket because a new motherboard design would be required anyway.