Slashdot Mirror
NASA To Get 10,240 Node Itanium 2 Linux Cluster
starwindsurfer writes "US space agency Nasa is to get a massive supercomputing boost to help get its shuttle missions back in action after the 2003 shuttle disaster. Project Columbia, a collaboration with two technology giants, will mean Nasa's computing power will be ramped up by 10 times to do complex simulations."
...but someone ought to tell them that Doom 3 runs pretty well just on moderately-new hardware...
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
10240x more dupes?
I want to delete my account but Slashdot doesn't allow it.
Well, I guess they're not using it to serve that webpage.
-m
#
# Modus Ponens
#
http://linux.slashdot.org/article.pl?sid=04/07/28/ 1427228&tid=163&tid=139&tid=103
Do the editors work for the USPTO as well?
// Agent Green (Ian / IU7 / KB1JQO)
// IEEE 802.3: All 10base Are Belong To Us
"are we gonna get another barrage of (insert slashdot cliche') posts" again?
Damnit!
About $7.2 Million.
Talk about a software tax!
"Rocky Rococo, at your cervix!"
This should help 'em convert feet to meters ... ;-)
Hulk SMASH Celiac Disease
This is great news for intel. They will double the number of itanics shipped in a single deal!
Hahaha, my comment is a dupe!
The system will have 500 terabytes of storage, the equivalent of 800,000 CDs.
In related news, the RIAA has filed a writ of discovery for illegal downloads of 'Major Tom' at NASA.
But I wonder if moving from a spreadsheet to a supercomputer simulation will make it any more likely that engineers with concerns will whistleblow to non-responsive management. This is a government bureaucracy problem, not a technical problem.
I'm rather mad at this idea, the system costs more than an opteron system, costs more to run (heat/power) and is slower. But it at least runs linux.
Also, why is the BBC the first news tidbit about NASA's new supercomputer?
...a Beowulf cluster of slashdot dupes.
pb Reply or e-mail; don't vaguely moderate.
So Hrathgar carries Hrunting into a bar. The bartender asks him "Why the long face", and Hrathgar cuts his head off with Hrunting singing, "A hrunting we will go, a hrunting we will go!"
*Rimshot
.. or a very good writer.
"They can also be modelled over a time period of weeks or months instead of over just a few days."
Ohh sweet, so then what used to take days now takes months?
And at one point in the article, it says "20 nodes" and then at another part it says "512 nodes." So like, what is it?
You know what, I don't even care.
- It's not the Macs I hate. It's Digg users. -
I can understand the BBC making this mistake, but slashdot?! I'm sure this was also noted in the dupe.
Democrats and Republicans only disagree about how to enslave you
http://www.sgi.com/newsroom/press_releases/2004/ju ly/supercomputing_ctr.html
Rus
Cheap UK and US VPS
Apple didn't give VT any computers, they paid for them because they were the cheapest solution.
if they'd gone with G5 Xserves they could have had 23,888 Dual 2GHz systems with 17.916 Petabytes of storage (assuming they just went stock on the high-end systems).
Okay and one question about the article. Was he saying 1000 Gb of RAM per system or 1000GB per system?
Does anyone know of a better site than /. to discuss tech related topics because, as the post I am responding to shows, /. users are getting dumber and dumber as time goes on.
Think of it less of a win for Itanium and more of a win for SGI Altix (that happens to use Itanium). The SGI Altix machines have a single system image with 512 processors (there are 20 of these clustered together). As far as I know, this is actually the cheapest and highest performing system that can use 512 nodes in a single system image. Other choices (which I'm not even sure scale to 512 processors) include Sun (slow), Power (expensive), and MIPS (SGI predecessor to the Altix - slower). Also, they are working on methods to increase single system image size to 2048 nodes, I believe an industry first. Some workloads just like running in single system images much better than on clusters.
As for Itanium vs. Opteron - the Itanium kicks the Opteron's ass in floating point. Since NASA is presumably going to be doing a lot of engineering simulations, good FP performance is highly desirable. Having 6 MB of cache per node probably helps the Itanium beat out the Opteron for large memory footprint workloads as well.
Basically, until Cray releases Red Storm (not sure if they'll stay in business that long), an Opteron system doesn't exist that can match the performance of the SGI Altix.
Finally, Itaniums are NOT "rediculously more" compared to the 8xx Opteron line (which is the Itanium's real competitor in this area).
Dan
It does floating point a lot faster than Opteron.
Why use itaniums? Because itaniums are very fast at floating point math, and have 9MB of cache. It's not a perfect CPU, but it's not bad. Nasa is more than willing to optimize their code extensively. (Yes the optimizing compilers ARE available, just not in gcc. Both intel and hp have very good compilers for ia64) The IBM power architecture is also a very good architecture, but they are also VERY expensive.
Mostly they use Itaniums because they are buying an SGI sollution. Nasa Ames has been a long time sgi customer. The cluster of itanium/linux altix machines is simply a kicker to their previous cluster of mips/irix origin 3000 systems, which replaced a cluster of o2000s, which replaced a cluster of power-challenge boxes. That's one of the reasons this purchase happen so quickly. All the physical/technical/knowledge/business infrastructure was in place.
If you read the sgi press release, they are also cutting nasa a huge break on the price to win the contract. It's about $2million each for those altix boxes including fibre channel cards, switches, and storage. I can't believe SGI is making any money on the deal.
where is SUN Microsystems?
:)
well someone had to ask
Don't Tread on OpenSource
The simulator originally ran on IBM System 360 mod 75's (serial numbers 1, 4, and 5). When I was working on it, the simulator was running on a IBM 3033 (370 architecture) machine running MVS, and had a hardware interface that attached 3 AP101's to the system IO channels. The shuttle hardware outside of the AP101's and environment were modelled in the 3033, even including the "slosh dynamics" of the fuel in the external tank. The simulator was written in 370 Assembler with macros for the programming control structures.
One of the funniest things about running the simulator came out of the major failure tests. The simulator had a distinct "abend" that indicated that the vehicle had a position that was below the surface of the earth.
It makes you believe this supercomputer is made out of commodity components.
That's blatantly false.
The SGI systems are highly proprietary equipments that provide very large bandwidth between the nodes, extremely low latency and tight integration. They're not regular Beowulf clusters. They really are single systems with hundreds or thousands of CPUs, all of them running the same single instance of the OS (as opposed to typical clusters which run one OS instance per node).
Because of the tight integration, the software does not have to obey the same constraints as when running on commodity clusters. Especially the requirement for total parallelization does not stand anymore.
Therefore, problems which cannot be translated into 100% parallel algorithms, and therefore do not run efficiently on commodity clusters, are easily tackled on SGI supercomputers.
That's why they can charge a high price on their systems - because they can solve problems that are not accessible to "normal" computers.
That being said, the system at NASA is indeed a cluster, but it's a "small" cluster (a handful of nodes), each node being a supercomputer with hundreds of CPUs. It's a hybrid that provides the best of both worlds.
Because it's a 20-nodes cluster, each node being a supercomputer with 512 CPUs.
u ly/supercomputing_ctr.html
The article was written, unfortunately, by a rather clueless journalist. Here's a link to the proper information:
http://www.sgi.com/newsroom/press_releases/2004/j
The "firmware" (the equivalent of BIOS) they have on the Altix is pretty damn smart, it's like an OS of it's own. It can do diagnostics, and inventory and a truckful of other things.
Powering up a huge complex beast such as an Altix is no easy task. You need lots of "intelligence" at the hardware level to do that.
There is a limit to what computer power can do for you. I'd rather see the money being spent on human resources: people who know what they're doing. There's an old saying in the business world, I wish I knew who first said it, "for any technological problem, the limiting factor is never technology, but rather, human resources." In other words, if your technology has problems, throwing more tech at it is unlikely to solve the problems. Only more human intelligence applied to the situation will improve things.
Having the fastest supercomputer in the world won't help you one bit if nobody thinks to run a simulation of what happens when a chunk of foam blows a hole in a wing. I keep thinking about Frank Borman's statements to the Apollo 13 Commission, he said it wasn't a failure of technology, it was a failure of imagination, nobody ever imagined there could be a problem. Computers have no imagination. They give answers, but nobody's asking the right questions.
How did this comment got moderated "informative"? There's definitely something wrong with the moderators today.
SGI Altix uses the Intel compilers. They're pretty damn good on IA64. They're available today.
Also, the massively parallel software is already up'n'running. NASA has been using for decades SGI supercomputers - traditionally it's been the MIPS/Irix architecture. A while ago, when SGI told NASA that they were going to migrate to Intel/Linux, NASA simply recompiled their software to Linux, which is not too difficult, since Irix is pretty much standard Unix (i did some porting from Linux to Irix and often the software simply compiles with no change).
Also, Altix systems are essentially the same hardware architecture as MIPS-based SGI Origin with the exception of the CPU (and a different OS on top), so the differences are really not that big; it's just the transition from Irix to Linux.
They better pay their $7,157,760 ($699/CPU) in SCO tax or McBride is going to be stomping around saying "NASA is screwing us!"
"I bow to no man" - Riddick
Imagine a Beowulf cluster off... oh, wait...
The issue has become space to put the machines. As it is, the pre-existing supercomputers are being moved to other rooms and there is barely enough space to accomodate the 10k as it is. Many supercomputing facilities have a similar problem. There are not many rooms that have the environmental controls needed to run such massive systems.
As for the comment on making it 11x faster - the other systems serve a different purpose (customer base and funding source)... and they were moved to another location to make room for Columbia.
On a cool note, it looks like they are filming the building of this system so we can see one of those time-lapse videos.
NASA is not the sole user of the system. Anyone within the U.S. can use it. We support many university projects that require the use of supercomputers. This purchase is a benefit to the entire country, not just NASA.
It is most unfortunate that people are not aware of all that NASA does for them. A majority of all research projects are in collaboration with industry vendors, universities, non-profit organizations, scientific corporations, and so on. There are few that are specific only to NASA. The range of customer database is wonderful and there is such variety in the areas of research (not just aeronautics and space technology, but biology, earth science, nanotechnology, optics, and so on). We all help each other to advance our knowledge, and computers like this make it a lot faster.
The Columbia disaster was not due to a lack of computing power, but rather to a culture of denial. The failure of mid-level and senior management to listen to their people prevented any action being taken until it was too late. In a way, this mirrors the broader American culture of the late 20th and early 21st Centuries, typified by a complete refusal of individuals - particularly but not exclusively individuals in powerful positions - to take responsibility for their own actions, inactions or failures of any kind.
NASA does not care about money. It's US taxpayers' money
Thanks for the troll post.. you're a wonderful example of how uninformed citizens can be. FYI: The government defines NASA's budget each year, so there is a very high concern for money. It takes months upon months of civil servants fighting for funding out of that money pool. There are a lot of research programs, and not nearly enough money to fund them. Particularly, in the case of Columbia, there were massive layoffs to fund this. I'd like to see you make your statement to all those who now do not have jobs because of the lack of money (many were needed operational engineers, not just research staff). It's sad when people lose their jobs over something like this, but it did allow something good to happen. It's unfortunate that arrogant fools are blind to such politics.