Sandia, Compaq, and Celera To Build Petaflop Machine 113
Fact-o-matic writes: " Compaq, government weapons facility Sandia National Laboratories and genetics researcher Celera Genomics are teaming up to build a petaflop computer -- one that will process 1,000 trillion operations per second.
To listen to an audio playback of today's press conference announcing the project, Celera has set up a phone-in recording: call (800) 642-1687, and enter the conference ID: 818790
You can read the joint press release or the
Compaq press release"
Alternate Number (Score:1)
Re:Beowolf Cluster (Score:1)
Re:Petaflop machine (Score:1)
Have we suspended Moore's law? The ASCI Red (currently 2nd on the Top 500 list [top500.org]) has 9632 processors, and it broke the Teraflop boundary with plain ol' Pentium Pro processors--1996 technology. (It's since been upgraded) Now, say this vaporware computer shows up 9 years later, that will give processor performance roughly 2^6 of an improvement. With 10-20K processors, that's roughly 64-128 Teraflops, not a Petaflop.
Now, 100-ish Teraflops is a lot more realistic than a petaflop, especially since this article [zdnet.co.uk] quotes a figure of 100, not 1000 Teraflops.
A Reality Check (Score:3)
Yes, it's another Linux win ... (Score:5)
"The Sandia labs and Celera Genomics Group in Rockville, Md., plan to work together on the project under a joint research and development agreement, with Houston-based Compaq as their technology provider. The planned system will be built around future versions of Compaq's AlphaServer SC supercomputer line and is being designed for use in complex applications in the fields of computational biology and life sciences...."
"The prototype supercomputer will likely use 10,000 to 20,000 of Compaq's Alpha processors and is being budgeted at $150 million in current costs, according to Blake. He added that the first system could eventually lead to the development of a so-called "petacruncher" -- a machine capable of 1,000 teraflops -- by the end of the decade."
Re:I wonder... (Score:1)
Re:Petaflop machine (Score:3)
projects like distributed.net and seti@home are perfect for distributed systems b/c the data analysis can be easily split into manageable sets which can be passed off to willing participants.
If you are referring to something more along the lines of a beowulf cluster, then you should take a look at 3. of the beowulf faq [dnaco.net].
In order for a cluster of machines to make a difference, that application has to be built for such a thing.
hmm...wait a second, did I just respond to a troll?
-Peter
Re:Yay! (Score:1)
Re:Yay! (Score:1)
It seems every one that parades an interest group on slashdot makes wild generalizations about the "evil" out there, evil business, evil government, evil science, evil technology, evil lawyers, evil this, evil that, trying to promote a serious interest on slashdot is worthless.
Frankly all this activism is getting old on me.
I will note the remark about peta flop is a little out of line, but they try to make themselves out to be the most extreme of the extreme groups and seem to get confused when no one agrees or understands them.
The one thing that gets me is this:
millions of innocent creatures die every day
Do you even have real statistics to back THAT claim up? Millions of creatures die every day? I think that figure is a huge streatch even if it was replaced with every year. Come on, admit it, it is sensationalism.
I am against animal abuse but I am also against sensationalism.
what a waste of phone time (Score:2)
how big was their phone bill for this?
http://www.linuxtoday.com has the link. (Score:1)
Looks like it runs Linux. (Score:2)
Is this the first time? (Score:1)
you pay for it in drugs (Score:2)
segment that can afford $100 million computing
initiatives. Many of those fancy new drugs cost
hundreds of dollars per month per prescription.
government computer welfare (Score:2)
with the funds to push computings' edge, i.e.
that is systems over $20 million dollars.
That buys 10 terflops now and by 2010 a petaflop.
Now and then an industry will be doing good enough
to buy the large machines- drugs, oil, Hollywood,
but that is transient. The government has been
supporting advanced computing since computers
were invented around WWII, instutionalized in
DARPA and the National Labs.
first petaflop conference in 1994 at Caltech (Score:2)
when the teraflop barrier was surpassed.
Moore's Law predicts a factor of a thousand more
or less in 15 years. The issues were whether
conventional hardware and software development
would make this next jump of a thousand,
on radical new inventions would be needed.
Currently it looks like the existing trend
should squeak by, but physics will impede the
next jump of a thousand to exaflops.
Assembling DNA Fragments (Score:2)
The problem with this is that the human genome has lots of repetitive sequence which make it hard to identify what fragment goes where.
Compare to the government-funded approach of first breaking up the genome into manageable chunks of known location and size, then break those up into smaller extra-small chunks and then re-assemble those. Then you know what you've already done, and where you've got gaps.
Obviously taking fragments from all over the genome and assembling is going to take a significant amount of processing power, much more than it would take to just do one smaller manageable chunk of the genome.
Government Funding/Conflict of Interest? (Score:2)
Now correct me if I'm wrong, but I'm seeing something wrong here.
So technically not only are our tax dollars funding the government-funded Human Genome Project, we are also paying a national laboratory to help develop a supercomputer for commercial interests, correct?
Sandia Lab and the Human Genome Proejct are both run by the US Department of Energy, so I certainly seem to see a conflict of interest here... anyone else?
I'm sorry (Score:2)
... but Imagine a beowulf cluster of these...
Re:Yes, it's another Linux win ... (Score:1)
Re:Yes, it's another Linux win ... (Score:2)
Linux support for the high speed interconnect that they use (Myrinet)
No, but because of jokes like that... (Score:1)
---
Bubblegum Crisis reference. (Score:1)
(Bubblegum Crisis is a Japanese animated series, about chicks in power armor fighting the Boomers; an overrated one, IMHO)
---
Why would anyone ever need more than 640K? (Score:1)
Because it's possible.
After they've done it, people will have no problem finding ways to exploit it to the maximum.
It's always been the same with any new advance in computing technology.
What is the real use for this computer? (Score:2)
Geome matching needs a nice long word (65k bits?) registers and a fast word sized barrel shifter/comparator. That is what is need to search for sequences that appear in several places.
Cray computer company (not Cray Research) were building a device like this for the nice spooks at the NSA when they decided they didn't need it anymore an cancled the contract. That box looked like it was designed from the ground up to do research on very long bit streams (say RSA type keys) but they canceled the project after paying something like 90% of the box and the bits that had been built were distroyed. Too bad that machine didn't see the light of day, it would have been great for finding patterns in DNA sequences.
Re: (Score:1)
Re: (Score:1)
Re:Yes, it's another Linux win ... (Score:1)
I have no doubt that over time, that gap will close significantly when things like page coloring and other performance optimizations are built into the Linux kernel.
One thing Linux has over Tru64 is that there's alot more drivers for alot more peripherals. Myrinet, Dolphin, 1000Mb ethernet, etc, all work just fine on Alpha Linux. All are used, depending on the application, by folks in the HPC (high performance computing) world.
probably the first time... (Score:1)
Peta or Tera? (Score:1)
------------
What's it going to do? (Score:2)
Does that phrase scare you as much as it does me?
I assumed it would be for nuke simulation till I saw that. This strikes me as somehow worse.
The real Eight Star misses Technocrat.
Re:Looks like it runs Linux. (Score:1)
Re:Why any operating system? (Score:1)
Then I got to thinking... why not Linux? They are already going to spend doubtless tens or hundreds of millions of dollars on it, so improving SMP is probably not too tricky. Additionally, there are at least two Real Time implementations for Linux that I know of.
They can tweak the hell out of it because they have the entire source code in a mostly unrestricted form. Sun's kernel is "free" but you have to pay for it after using a certain number of CPUs. Also, the Solaris kernel is designed to be constantly defragmenting memory. That's why it runs so slow (which is where the "Slowaris" moniker comes from). They aren't running a high-end Web server sitting on an OC3, so this constant memory defragmentation is probably detrimentatl; the data structures they will be working with will be unbelievably huge, and if the kernel spends time trying to optimize allocation, it's going to choke down to a halt.
So yeah, I think it was a wise decision to use Linux. Since they are not going to be releasing binaries, they are under no obligation to provide the kernel patches, but if they're nice they will. Linux could really use better SMP support. As far as the heaviness of the kernel, you can build a very lightweight kernel by just turning off all the stupid crap in 'make menuconfig'. :) - They can also turn off a lot of stuff by manually editing the kernel source, as they doubtlessly will.
Re:Yay! (Score:1)
Re:Government Funding/Conflict of Interest? (Score:1)
Re: Yay! (Score:1)
Re:Yes, it's another Linux win ... (Score:1)
Think we can /. a phone number? (Score:4)
Re:Petaflop machine (Score:3)
We can double the transistor density whenever we want. Economic factors will effect it as will government research policies.
There is no need to exceed Moore's law, you could just increase the number of processors. However, if we need to increase faster, we can, we just pump money into engineering research.
There are laws of physics - these can't be broken (although the law could be changed if we had a problem)
There are laws of society/traffic - these can be broken but you can get in trouble
There are historical trends with cool names that include law... these can be broken whenever someone is in the mood too...
Re:Petaflop machine (Score:1)
Re:Petaflop machine (Score:1)
Re:And another thing. (Score:1)
Re:Petaflop machine (Score:1)
Re:Petaflop machine (Score:4)
> do that can't be done by a distributed system?
Not every possible computation can be made distributed without major performance loss. Remember, this baby IS vastly parallel (10K-20K processors), but the inter-processor communications are way faster than any network, and some problems that can be parallelized aren't easily distributed.
Also, i doubt you want nuke research done on the same set of systems running SETI@Home :-)
-DVK
How fast is that? (Score:1)
But of course the entire one hour interview lasts only one one thousandth of a second.
:-)
Re:Yay! (Score:1)
Re:Yay! (Score:1)
Much as I hate to join in these silly flame pickles, I don't think (as your login suggests) that pet owning is slavery. Indeed, many pets have shown a commensalism with man over many hundreds of years (dogs, cows). What would have happened to the stupid cow if we hadn't found a use for it? Well, they wouldn't be nearly as abundant today, that's for sure. They gauranteed the survival of their species by being a use to us. You'll find the same things elsewhere, such as lichens.
Re:What's it going to do? (Score:3)
It's more that both of them need serious computing power than that they're thinking about biowarfare. Bioinformatics has the potential to use truly monstrous amounts of processing power. Assembling a genome that's been shotgun sequenced is going to require serious computational horsepower, and Celera wants to start pumping out genomes left and right. Annotating the things is also going to be pretty brutal, although with a few genomes as roadmaps this may be a bit less trouble. Any way you slice it, though, you're talking about needing massive number crunching power, and nobody knows more about that kind of things right now than the nuke simulation boys at Sandia, who are currently being encouraged to branch out and do more than weapons development. It's a reasonable match.
Other famous shortsighted comments: (Score:1)
Wow! I had to double check I was still reading Slashdot! That was one sentance I NEVER thought I would read here... Let's take a brief trip down memory lane:
"The world probably won't ever need more than a half dozen computers or so to fulfill its computing needs."
Or how about,
"Computers will never need more that 640KB"
My point is, decades from now, even a petaflop will probably look tame.
Re:Petaflop machine (Score:3)
Incorrect (Score:1)
This is incorrect. Research has shown that other animals, whales for instance, have a similar understanding and appreciation of music. I point you to this article:
http://www.nytimes.com/2001/01/09/science/09MUSI.
Indeed, some species have advanced past "fscking and eating and sleeping."
Quake on that baby.... (Score:1)
My thoughts ... (Score:1)
I wonder how long it will be before there are men standing on street corners with signs, "Will code for food" or "Will hack for clothing".
The economy is cyclical. It's going to happen. And when people have to tighten their belts, you better believe all these techno-toys are going to be one of the first things to go. So hold on while you can, find a job that isn't going anywhere (government is a great place for that kind of stuff) and hope for more projects like this.
- W
Re:Yay! (Score:1)
Re:Redundancy check (Score:1)
Yeah, yeah...petaflop (Score:1)
Re:What's it going to do? (Score:1)
Listen. Sandia is a DOE lab, and the DOE has its hand in all sorts of pure scientific research [energy.gov] that has nothing to do with the military. Sandia is only partly a weapons lab and pure research is exactly the purpose of much of the DOE budget.
That's a lot! (Score:1)
<Dr. Evil Voice>
</Dr. Evil Voice>
Lights out (Score:1)
When it's not crunching numbers at 1000 Tflops, will they at least shut down unused parts of the cluster to be wake-on-LAN?
-ez
Compaq fools em again (Score:1)
Re:Think we can /. a phone number? (Score:1)
how stupid of Celera! (Score:1)
Sarcasm aside... that seems a little backwards for someone working on a fricken' PETEFLOP machine.
E.
www.randomdrivel.com [randomdrivel.com] -- All that is NOT fit to link to
Decompiling DNA, etc (Score:3)
Problem is, the doin't know what the genes actually do, for most of them.
So the extra computing power has to go into sorting this out, and figuring what they mean, the grammar of the genes.
The only thing I can think of, that would be like that, would be the old translating of the rosetta stone.
Now that was a political text. Now what if the text had actually be a discussion of the subatomic particle reactions that take place in a matter/antimatter reactor? The task of translating for Napoleons archeologists would have been much harder, even *if* the ancient greek had been in "clear text", because archeologists did not know higher mathematics, etc. Probably, it would have been seen as a *really* obscure alchemy or religious text.
So now we have a similar task. Knowing the letters and a few "words" of the genome does not mean anything like knowing the design principles that are incorporated into a strand of DNA. DNA is all compiled coded, and we are trying to manually de-compile it. Then reconstruct the source.
Re:Is this the first time? (Score:1)
Epoxy (Score:1)
That phone number... (Score:2)
just a thought...
Re:Yes, it's NOT another Linux win ... (Score:1)
The Computer World article you link to is just a summary of Compaq's press release (read the first sentence of the article). The real press release does not say that the new super computer will use Linux as the OS. In fact, from the press release: Compaq and Sandia will collaborate on the development of system hardware and software. Both have extensive experience with supercomputers based on Alpha. The one and only mention of Linux in the entire press release: Sandia currently operates the most powerful Linux-based supercomputer in existence, Cplant[tm], which employs more than 1600 Alpha processors. As you can see, it is a previous (and much-less powerful) supercomputer that was running Linux.
Medicine, not bioweapons (Score:1)
and another thing... (Score:1)
Wrrrrrong! (Score:2)
petaFLOP == 1,000 trillion FLOATING POINT Operations Per Second.
It turns out that the author of the article linked to uses the term to mean 1,000 trillion ops/sec too. I think this is correctly called 1,000 TIPS (trillions insturctions per second), although IPS is usually not that commonly quoted.
I see an IP lawsuit around the corner. (Score:4)
---
...forget the obvious... (Score:1)
If I'm really excited about this... (Score:1)
Re:Yay! (Score:1)
Re:You can bet . . . (Score:1)
Re:CHANGE THE FUCKING POLL ALREADY (Score:1)
Re:Government Funding/Conflict of Interest? (Score:1)
Why any operating system? (Score:3)
Old News - IBMs "Blue Gene" First (Score:2)
Re:Yes, it's another Linux win ... (Score:1)
-- The Unix Programmer's Manual, 2nd Edition, June 1972
Re:You're wallowing in your own error. (Score:1)
Read a little closer next time.
Redundancy check (Score:3)
And that would be one quadrillion. Come on, don't be so afraid to say quadrillion. I know you can.
Phone number? (Score:1)
Re:Eating meat is NOT good. (Score:1)
Although people do not require meat, they do in most enviroments require the nutrition provided by meat. I make that distinction because beware, your health is in serious risk of you disregard this fact. We have the adaptions of an omnivore, our bodies can process meat so I think our bodies are made to take advantage of it. You must find substitutes for meat nutritionaly. I'm not a vegitarian but I like to eat and cook vegitarian. Try nutritional yeast, its very tastey and has nessisary vitamin B12 (If I remember correctly).
"Think of the sentient beings that died, and you are eating every time you put a steak in your mouth. I heard they eat horses in France. If they tried to eat one of my polo ponies, I would be extremely pissed off. "
Your pony is your property. It probably enjoys this status... I've heard other forms of life eat animals. What of a lower sentient form like a shark eating a dolphin? What of a Whale eating plankton? What of a lion eating a human? Its all about nature and survival. Life is a limited resource all creatures try to make the best of it. We can observe this. This is something of interest to life that can contemplate it.
"No, there is no excuse for eating meat. Horses don't need meat, cows don't need meat, monkeys don't need meat - and neither do humans, in fact, no living being needs cooked food at all, for that matter."
Its not an excuse, its a life style. Ask any human eating lion...
ooops (Score:1)
Re:Redundancy check (Score:1)
--That's not 1000 trillion, it's 100.
That's where you got 1000 trillion from. It never said 1000 trillion.
Oh no... (Score:1)
Voice telling me not to post fading...
fading...
... fading...
gone =)
A petaflop computer, running linux...
IMAGINE A BEOWULF CLUSTER OF THESE!!!!!
Sorry everyone, it's late and I saw that number and almost dropped a load when I saw it. Let the flames begin ;)
Re:Think we can /. a phone number? (Score:1)
(p.s. couldn't seem to (be bothered to) find the article).
Re:Think we can /. a phone number? (Score:1)
1-800-888-3999, and select option 7
Still works
Petaflops for Gnome (Score:2)
Re:Why any operating system? (Score:1)
Not that that matters much, considering the development costs...
Re:I'm sorry (Score:1)
Celera Genomics? (Score:2)
Yay! (Score:5)
I'm glad all that crazy animal rights group flopped!
Imagine the possibilities (Score:1)
This is great!!! MPIS for meesss!!! (Score:1)
Android design? is this another joke? (Score:1)
Re:Petaflop machine (Score:2)
I'm sure glad Intel doesn't have a hand in this. Holy Global Warming!
I'd rather be a unix freak than a freaky eunuch
And another thing. (Score:2)
They want the government to determine the course of genetic research, or at least the course of the distribution of information produced by genome research.
They don't care about the petaflop. Petaflops will be designed in many forms. And one that takes this many processors is not industrially replicable, anyway. So Compaq gets some advanced r&d but very little salable out of this collaboration.
Truck drivers... (Score:2)
I'm what you might call a formulations development chemist. That means, when a client comes to us with a drug, I figure out how to get the proper release profile from the depot formulation that is to be used. That means I'm required to know organic chemistry, pharaceutical chemistry, pharmacokinetics, and polymer chemistry. Oddly, it does not require driving a truck.