IBM's Blue Gene powered by Linux

bigjnsa500 writes "Linux will be the main operating system for IBM's upcoming family of 'Blue Gene' supercomputers--a major endorsement for the operating system and the open-source computing model it represents. Blue Gene/L, the first member of the family, will contain 65,000 processors and 16 trillion bytes of memory. Due in 2004 or 2005, the system will be able to perform 200 trillion calculations per second. Lawrence Livermore National Laboratory will use the system for performing nuclear weapons simulations." Blue Gene has been announced for some time, but it's cool to see how it's shaping up.

Comments

SCO Linux

[petecarlson]

Hmm, I wonder why they chose to use SCO's OS. You would think with all the lawsuits they would try to stay away from SCO's software...

ASCI Red Storm runs SuSE Linux

[Boone^]

...and it's got 10,368 2 GHz Opterons. (link)

ASCI Red Storm google search

Re:yeah but...

[mister_tim]

Maybe not - but given IBM's history, it might be great at playing chess.

Or Tic-Tac-Toe, given the nuclear weapons simulation angle.

in other news

[t0ny]

In other news, Levi's has announced a lawsuit against IBM, citing the name of the server line could confuse their customers.

Zowie

[Anonymous Coward]

Blue Gene/L is expected to operate at about 200 teraflops which is larger than the total computing power of the top 500 supercomputers in the world today.

It should be noted

[wmaker]

It should be noted that these super computers won't be for sale... IBM simply leases the cycles, you pay based on the cycles you use every month.

How many apples is that?

[goombah99]

Wow 650,000 processors. what's that in equivalent g5 flops, say 4000 or so :-P

kidding aside, are these based on the novel IBM design for having small clusters of wimpy processors sharing sections of memory. The concept being to have each processor running slowly, almost stalled waiting on a memory fetch. (while seeming stupid at first glance, its really diabolically clever since now you can junk all the long pipelines and branch prediction stuff: every single byte that comes from memory will be used by some CPU requesting it, thus you minimize the memmory buss buttle neck that is, ultimately, the limit on most processing).

if this is that design then that 65,000 processors indeed may not be quite as much computing horespower as it sounds. it might indeed be comparable to a smaller handful of G5s.

or maybe i'm full of crap.

Re:How many apples is that?

[ocelotbob]

Uh, IBM makes the G5, or rather, the PPC970. I think they of all people would know whether or not the processor is suitable for the task at hand. Don't you agree?

Re:How many apples is that?

[tychay]

You say:

The g5 supercomputer mentioned on slashdot before never performed what it originally claimed.

What was the claim? The only bogus claims I heard regarding the Terascale (G5 cluster) were:

1. Ignorant people taking Rpeak and multiplying by the number of machines and,
2. Wired taking an accurate claim of the clusters performance on 128 CPUs and extrapolating it to 2200. In the article, the manager of the top500 noted that the G5 cluster might take #3 and contend for #2.
3. A whole bunch of FUD from people like you who have some reason to wish the people working on this project ill simply because they chose Macs to do it.

Then a New York Times report using old data reported 7.1 teraflops Rmax--enough to put it at #3 on the old list and #4 on the new--NYT forgot to mention that there have since been three new clusters that made the top 10, one of which slightly edged out the Terascale.

Of course, by the time that was reported, the figure was revised to 8.3 Tflops and now, officially reported (both on the current Top500 and by the head of Terascale) as 9.555 Tflops (60% efficiency) with the stipulation they could probably get 10% more. A pretty umapproachable #3 spot in the Fall500 and the first sub-$100 million dollar system to break the 10 teraflop mark.

Go look at the current benchmarks, where are the Pentium clusters that are above it? Where are the Itanium clusters above it? Where are the Athlon clusters above it? Oh, I'm sure there will be some (probably in the Spring2004 500), but where are they all right now? How much do the current ones on the list cost (answer: no less than $30 million). Sounds to me the wishful-thinking, poor-reporting Wired and the Mac zealots were closer to the truth than FUD-meisters and the anti-Mac zealots.

The most efficient top 10 supercomputer right now is also the most powerful: The NEC EarthSimulator at about 80%. I'd imagine we should expect a 60-80% efficiency from the big budget Blue Gene/L. And in my book there is nothing wrong with the current 60% efficency of Terascale--anyway it probably says a lot more about how good Infiniband is than it does about how good the Mac is.

But the writing is on the wall. There is nothing special about the the 970 (G5), Virginia Tech could have done the same thing with an Opteron or Itanium2--it would have taken more processors and cost twice as much: ~$10 million best offer for the systems as opposed to $5.7 million list price paid for the Macs (subtracting $1.5 million for the Infiniband cards, routers, and cabling).

The take home point is not that they did it with Macs or Mac OS X instead of (your favorite CPU) and Linux. The take home point is: these guys built a top 10 supercomputer in a fraction of the time (months as opposed to years) at a fraction of the cost (<$10 million as opposed to >$100 million).

Yes, like the Crays of the old days (and today) there will always be those who need something like Blue Gene/L and IBM is happy to supply them. But a whole new generation of supercomputers will be built on-demand and out of commodity PC hardware and a good set of software running on an OS that doesn't charge for all the CALs. Right now the 970 is easily the best performer for LinPak. So much so, they can pay educational list price which included such worthless features as an Apple-tooled case, overpriced RAM, gigabit cards, and Radeon graphics cards, firewire, usb2.0, digital audio, iTunes and other iApps, and a OpenGL based desktop. Since the 970 is made by IBM, I'd hazard a guess that IBM would be happy to supply these people too. Whether they choose to run Linux, MacOS X, or something else.

Nuclear Weapons

[jonhuang]

Somewhere, there's an open source developer who's just realized that his work is being used to the development of nuclear weapons. All jokes about derivative works aside, I think it's a good time to consider the implications of this.

Re:Nuclear Weapons

[Shakrai]

Somewhere, there's an open source developer who's just realized that his work is being used to the development of nuclear weapons. All jokes about derivative works aside, I think it's a good time to consider the implications of this.

And Nobel discovered that dynamite could be used to kill people as easily as it could be used in mining or construction. What's new?

Re:Nuclear Weapons

[cgranade]

As opposed to if he charged for it, in which he would have just made blood money, and implicitly accepted the use of the technology. In this case, he/she can't say that s/he endorses the use, because they released the technology from their hands. To be sure, it is still disturbing, but not in the same way...

Re:Nuclear Weapons

[Shakrai]

To be sure, it is still disturbing, but not in the same way

True enough. But rest assured for everything you create with good intentions, somebody out there will find a bad use for it. And somebody like SCO will try to make money off it....

Re:Nuclear Weapons

[davejenkins]

Somewhere, there's an open source developer who's just realized that his work is being used to the development of nuclear weapons. All jokes about derivative works aside, I think it's a good time to consider the implications of this.

Somewhere else, there's an open source developer who's just realized that his work is being used to power machines for Doctors without Borders, the Red Cross, a number of innercity churches and rec centers, and hospitals.

Yet somewhere else, there are soldiers testing out new battlefield computers that run open source, and those machines are more stable, which means more safe, which means one more son doesn't come home in a body bag.

Honestly, does that sand around your head taste good?

Nuke simulations?

[gumpish]

Why are computers still being used for simulating nuclear weapons tests?

Are they trying to pack more megatons of destructive force into each warhead? Don't the major world governments have enough quantity to preclude the need for more powerful units?

Or are the tests run to design "safer" and/or more localized implementations? (Awww, looks like Big Brother has a soft spot after all...)

Re:Nuke simulations?

[Shakrai]

Why are computers still being used for simulating nuclear weapons tests?

I should have linked this in my other reply. My bad. Information on so-called "low-yield" nuclear weapons for the morbidly curious.

"I know not with what weapons World War III will be fought, but World War IV will be fought with sticks and stones." - Albert Einstein

(And we thought we were past all of this...)

Re:Nuke simulations?

[cranos]

I'ld much rather they were running comp simulations than real tests. "Look mah, no hands, teeth, hair, nails...."

Re:Nuke simulations?

[RevRigel]

The primary use for these simulations is to verify that, based on the current state of maintenance/decay of our current stockpile of nuclear weapons, how well those weapons are going to work. Aside from the obvious issues with the active ingredients decaying over time, there are other issues with materials that must be simulated. It's not necessarily used for new weapons development, although that is one use.

Re:Nuke simulations?

[goombah99]

Why are computers still being used for simulating nuclear weapons tests?

Historically, the modern reason for computer simulation of nukes is to put a stopper in the nuclear proliferation genie. The logic is convoluted but sensible. The idea is that first you get a test ban treaty. Second, You offer economic and power production aid to all countries that dont develop nuke engineering or let you control their plutonium bearing nuke waste.

this creates a situation where nuke weapon engineering has to be done either in secret (since there no civial reactor technology to produce plutonium) or if done overtly, they still cant test their weapons. Neither can we.

this leaves everyone in a delightful position of 1) not being perfectly certain their nukes will work when delivered. thus they are not good offensive weapons. (imagine what would happen if pakistan launched on india and it were a dud).
2) yet they still make good defensive weapons since even though its not tested it doesn;t mean it wont work.

which is sort of nice. it discourages both developement and first use. world is MAD but better off.

Unfortunately the US would never go for this if they did not have a way of testing their own weapons. So they do it in silico rather than in nevada. This allows us the political will to go through with this. a better world results. THe clock is ticking. we know the weapons will work now but they are aging.. will they work in say ten years. THis is where computer simulations come in. within ten years we should be able to model nukes and nuke aging on one of these machines at a level that gaurentees our readiness. or maybe if this test ban thing works we can just scrap them all in ten years.

that was the plan. But now with about 30 countries with potential nuke development capability this plan maybe about to break down. thus we go to plan B.

plan B is we use these big computers to design new reactors that dont produce plutonium. We sell these to the countries. now they can have nuke power without creating weapons grade plutonium. Again every body happy.

except of course N. Korea.

Re:Nuke simulations?

[Malc]

That goes on the pile with those other questions like why do the US and Russia still have thousands of missiles pointed at each other a decade after the cold war thawed? Why is the nuclear briefcase still following the President around and why are missile pointed at Russia still ready to launch with 2 minutes notice? Wasn't there an issue a couple of years back where the Russian President activated his nuclear briefcase on what turned out to be a false alarm?

I guess if you're going to maintain a nuclear "deterrent", you have to keep it up to date. Especially if your government has a nuclear first-strike policy. The comprehensive no test ban means simulation is the only option left.

Re:Nuke simulations?

[jsse]

Are they trying to pack more megatons of destructive force into each warhead?

Nope, they're trying to create a massive War-Sim in which nation-heads could raise war for a real deal.

"I'll nuke your a$$ unless you inflat your yuan."
"I don't fear your but I don't want to mess up the houses and railroads I spent three months on. You can get 6 yuan for 1 and you must neutralize a warhead in reallife as part of the deal."
"too late butthead, it has been launched already.Frankly, I just want to see how it explodes"

I can dream can I? :)

Re:Nuke simulations?

[rowanxmas]

So they do it in silico rather than in nevada

This is the best latin quote I have ever seen.

Um, maybe IBM should concentrate on making money.

[Rotten168]

Heh, while I realize that innovation has always been important for IBM and making money, it seems like they're trying too much to innovate and not enough to capture market share. It's like they're coming out with all these great inventions but they're pursuing pure science and not having a profit-making strategy.

Of course I realize that I'm probably wrong in some way but this is just how it seems to me.

Wait a sec..

[Epistax]

The IBM research team is currently running a large Linux cluster to simulate Blue Gene.

So then why don't have we have the simulation of Blue Gene run a simulation of Blue Gene two, and that run a simulation of a quantum computer, and that run a simulation of Deep Thought? Then that can run a simulation of the rest of the universe.
Then the two will bicker and argue about who's real, whom created whom, and millions of Matrix freaks will yell "I told you!!!" to those who have ridiculed them so many, many years.

Re:Um, maybe IBM should concentrate on making mone

[lucabrasi999]

not having a profit-making strategy.

Ahhh, you are forgetting the army of overpriced IBM consultants that you'll have to hire to install the thing.

Is this news?

[phch]

The referenced article is dated October 2002. Is this a mistake, or is this old news?

Anyhow, going to the Blue Gene web page, there is a document dated Nov 2002, an overview of BlueGene/L. An excerpt:

The approach we have adopted is to split the operating system functionality between compute and I/O nodes...

The compute node operating system, also called the BlueGene/L compute node kernel, is a simple, lightweight, single-user operating system that supports execution of a single dual-threaded application compute process...

I/O nodes are expected to run the Linux operating system, supporting the execution of multiple processes. Only system software executes on the I/O nodes, no application code.

Re:yeah but...

[Da Web Guru]

Well, since it will be running nuclear weapons simulations, they might get around to simulating the half life of plutonium...

Beating plowshares into swords

[Ralph Wiggam]

When did Blue Gene change to a nuclear simulation computer? Last I heard it was for protien folding and DNA research, which is why it's called Blue GENE. This way it's like the Utah Jazz.

-B

What video card to use with this bad boy?

[Anonymous Coward]

Do you go with ATI or Nvidia?

Good frame rate for Quake 3??

AA on or off?

VSynch on or off?

all well and good...

[pizza_milkshake]

but what kind of video card does it have? will have 65,536 monitor support?

Re:all well and good...

[mcrbids]

Well, there are 65536 digits spanned in 16 bits. Monitor 0 would be the first physical monitor, 65535 would be the 65536th monitor.

Who taught you to program?

So, where's the 0th monitor fit in here? Or, did you start at 1?

It's a blithe assumption by people that we start counting at 1. Ever wonder why it takes two digits to render just the first 10 counting in base 10, but only one digit to max out when counting in base binary?

If you count in base 2, you start at 0, as in

0, 1

not

1, 10.

It should then follow that

1, 2, 3, 4, 5, 6, 7, 8, 9, 10

really should be

0, 1, 2, 3, 4, 5, 6, 7, 8, 9

10 should start the next sequence since now we're an order of 10 higher than where we started, and we're using the extra place (the "1") to demonstrate this.

This is all standard number theory. The question is, where should we truly begin? For historical reasons, we start at 1, but for mathematical reasons, we really should start at 0.

Starting at 0, we'd have 65,535 monitors, which is the maximum value rendered by 16 places in base 2.

Who taught you to count?

-Ben

Re:all well and good...

[Hektor_Troy]

Starting at 0, we'd have 65,535 monitors, which is the maximum value rendered by 16 places in base 2.

Okay, I could understand 'interesting' but 'insightfull'???

Here's a nice calculation for you.

What is 2 to the power of 16? 2^16. Try all kinds of calculators.
How many times would you multiply 2 with itself to end up with an odd number (which 65,535 is)? It is fairly difficult to end up with an odd number when doing 2^y where y is an integer > 0.

Now, 2^15 + 2^14 + 2^13 ... 2^0 would yield an odd number (because 2^0 is 1), and incidently it would give you 65,535.

I'm guessing the guy who taught the parent poster to count actually knew what he was doing ...

SCO's Bill....

[Hanzie]

65,000 processors x $699/processor= $45,435,000. 45.4 million dollars.

Don't you just know Daryl's about to go apoplectic over all that money IBM is "stealing". Let's face it, he has to really believe in his private universe.

May he pop a blood vessel.

I think I know what IBM is thinking. . . .

[Com2Kid]

Perform some nuclear "tests" beforehand to ensure their next legal strategy against SCO will be effective. . . .

Those wacky propellorheads at Livermore!!

[morelife]

There they go using that unconstitutionally licensed OS again... with our tax dollars too. .. that will be able to perform a quadrillion calculations per second (one petaflop)...

That oughta give 'em the firepower to prototype the nuclear WMD that can surgically remove the state of Utah without bothering the neighbors.

about that nuke research-

[rritterson]

afaik, the research isn't on weapons development, but explosions research and weapons defense.(The more you know about the explosion the easier it is to design nuclear resistant bunkers and the like)

For those of you wondering why it takes 1 pflop to do such a simulation consider how much computing power it would take to follow each gas molecule in the explosion as it expands. They won't be able to get even remotely close to that precise, obviously. (6x10^23 molecules in 22 liters at room temp, so figure about 10^25 molecules to follow around)

Also, keep in mind that 70% of academic research dollars are defense related. (whether you like that or not, sadly)

What *I* Can't Wait For...

[philovivero]

When we get a supercomputer like this and the end of the article isn't "Some company will use this to find newer more efficient ways of killing people" but instead "Some university will use this to find ways of improving society at large."

I'm dreaming. I know.

Re:What *I* Can't Wait For...

[Copid]

Most of the supercomputers that people whine about because they support weapons development support a lot of other research as well. In fact, a lot less research would get done if we didn't have defense departments pissing away money on anything that might possibly have weapons potential.

Lots of wrothwhile stuff gets done on those machines, believe it or not. Just like lots of worthwhile stuff gets done at Lawrence Livermore Lab. They're famous for their weapons, but the amount of other research done there is staggering.

And in related News...

[Ieshan]

Michael Jackson has released a new hit single that denounces this upstart of a project as anything but his lover.

Re:Math makes my head hurt.

[Copid]

Remember, this is a simulation, not a simple matter of solving a system of equations. Most simulations involve evaluating the same set of rules over and over again over very tiny periods of time to get an idea of what happens over long stretches of time. Think of animation: Lots of drawings, each taking time to draw. Each drawing is based on the one before it, changed slightly. Eventually, you can construct a convincing simulation of motion.

In something like fluid dynamics, these programs are actually keeping track of particles bouncing against one another and updating the current state of the system over very tiny intervals. If you try to keep track of enough particles and make the time resolution fine enough, you're going to require incredible amounts of computing power.

Re:Um, maybe IBM should concentrate on making mone

[cmacmanus]

Wrong.

My Aunt and Uncle have been working for IBM since the "glory days" of computing, and through that I realized that IBM has a solid foot in the door. They provide servers/computers for hundreds of companies around the world, with the biggest being probably half or more of the current blue-chip corporations.

We're not in the dot-com era anymore, bud. :P

Cool name...

[glenebob]

They were going to name it "Billy Gene" but the name "Blue Gene" just "Beat It".

Yeah, 9th grade was like some sort of nightmare for me which seems to just live on and on...

Re:Um, maybe IBM should concentrate on making mone

[Rhys]

It's called government grant money.

Re:Sweet cluster

[ocelotbob]

We never left that era. There have always been $10 million+ supercomputers that fill up rooms. They went incognito during the 90s as datacenters, but they have always existed. It's just they take a bit of a different name now.

Bug report

[S.I.O.]

Dear Linus,

the kernel is becoming slightly unstable with more than 10 trillion bytes and 65000 CPUs, please try to reproduce the situation. See the attached memory dump file.

More corporate welfare...

[Anonymous Coward]

Lawrence Livermore National Laboratory will use the system for performing nuclear weapons simulations.

I commented on a similar previous corporate welfare handout where IBM was producing some software to mimic the human brain or some crap like that...to the tune of around half a billion dollars.

This is yet another such example...Lawrence Livermore National Laboratory is "operated by the University of California for the U.S. Department of Energy" This is yet another example of the public subsidizing hech tech industries, specifically IBM but it happens for others as well.

When are enough people going to stand up and put a stop to this bullshit so that we can use our money for much better use? Or better yet, when is the public going to be involved in deciding for themselves which projects get priority and how they are to be run?

And our government has the nerve to lecture others on how to run a democracy!

Finally!

[DownTheLongRoad]

A computer that can run Doom 3!

How long to crack 128 bit encryption?

[mcrbids]

Ok, assume I have a 128 bit key. How long to crack with a supercomputer this size? Anybody have a reference to mips->cracking time for something like this?

Just a thought...

Re:Um, maybe IBM should concentrate on making mone

[nobodys fool]

Maybe you should spent some time reading the works of Noam Chomsky ... but not those that every computer scientist knows of. He epxressed back in the 80ies that it was the public that paid IBM for developing computers in the 1950ies. One decade later, when a lot of money was spended and computers got profitable - of course, it was IBM alone making those profits - not the public.

It will be the exact same thing here. Do you really think IBM isnt paid huge amounts of money for this work? They are doing research here; they are making priceless experiences - and they are paid for it. The public is paying the research; the company will make the profit. It is always working like this.

Why do you think that any sane being would invest money for more and more powerful nuclear weapons? Because we need them? Bullshit. It happens because money is spent and some make profit for doing research. That is the reason. The only reason.

This is the price of freedom of knowledge

[heironymouscoward]

True, free knowledge can be used for evil or for good. For instance, a developer could place his open source work under a licence that specifically forbids specific uses (and this is something I did when I was younger, more idealistic, and less realistic). My early OSS works were not GPLd, but used a BSD-style license with certain conditions.
The problem with this is that you cannot simultaneously restrict and promote knowledge. As another poster has commented, everything we do as a society is interlinked: your taxes pay for guns and bullets as much as they do for medicine and books.
If a technology is truly free, it has no prejudices about who uses it. The GPL adds a second layer of freedom: it protects technology from being stolen and locked up again.
The OS developer who contributes to software used in the development of nuclear weapons will find one day that the nuclear weapons establishment has also contributed to the same software.
What I'm trying to say (and I worked all night on a stupid report, so my IQ is around 36 now), is that OSS is about the freedom of knowledge, and this flows in all directions: as much from the developer to the user as vice versa.

"16 trillion bytes of data"... Duhhhh.

[Bowie J. Poag]

I just love it when someone writes an article, and doesn't know hot to put it into words people can understand.. So they come up with this jackass "it's a million billion!" shit.

16 trillion bytes of data = Approx 15 terabytes.

What the hell is so hard about saying "15TB" ?

WTF? Nuclear weapons research?

[vandan]

If they're going to put so much money and so much effort into this, why do they have to research nuclear weapons? Surely we have enough weapons for everyone now. For fuck's sake. There's enough to wipe out all life on the planet hundreds of thousands of times over.

Why not research into harnessing different kinds of energy. Or search for a cure for cancer. Or look for fucking aliens.

But please. Not more fucking weapons. There are enough.

Why Linux?

[sharkey]

Noone wants Blue Screens on their Blue Genes!

Linux, sorta...

[jratt]

I recently started working at IBM doing iSeries Linux work, when I mentioned to someone that I had taken an MPI class in college. I found out later that he was doing BGL work and needed help, so now I work on this! IBM Research BGL Home

There are 65536 (2^16) compute nodes (CNs) on the system running a very small, from-scratch OS. There are also 1024 (2^10) I/O nodes on the system running a full Linux system (ZDnet article). The custom CN kernel is designed to look like linux, but is much smaller and written for a very singular purpose.

The system has a number of networks that link all the nodes together. The first is the 3-D Torus network, the point-to-point node connection topology. The asteroids game is a 2-D torus because the top connects to the bottom and the sides connect; a 2-D torus looks like a donut when connected together. A 3-D torus looks like a cube (3-D Mesh), but the sides are directly connected to the opposite end (it really requires 4 Euclidean dimension to draw well). This network only connects the 2^16 CNs.

The I/O nodes (running Linux) are connected by ethernet and then each linked to 64 CNs by the tree network. Unsurprisingly, it looks like a tree (for the people who actually know what a plant called a tree looks like, it is not like that).

Summary PDF

Becarefull with such "machines"

[smartdreamer]

Conversation logged at Blue Gene laboratory:

-- Roger, there's a bug with #21890, go check, STOP
-- Ok Houston. STOP
...
-- Houston, I lost my map. I can't find my way back. STOP
-- euh.. Roger, don't panic, we'll do somthing...
...
-- Houston, this computer is alive!
-- Yes Roger, he got self repare functions and evolution capabilities.
-- I s... I see something moving towards me!!
-- Can you repeat Roger?
-- mayday!.. heeeelp aarrGrgghh!! .......
-- Oh gosh, we lost one more.

Non Sequiter Re:Nuclear Weapons

[orichter]

It may be that Linux is currently being used to develop nuclear weapons, but this article has nothing to do with that. As the name implies, Blue Gene will be used for genetics research. Specifically, the protien folding problem, which in turn could help Geneticists to develop new wonder drugs without the current random trial and error methods they use. Imagine if we could simply plug in the code for HIV, run it through the computer, and custom design a drug to fight it. I'd think the developers of Linux would feel pretty good about that.

Blue Gene != ASCI White [Re:Nuclear Weapons]

[alacqua]

Actually, Blue Gene is being built to simulate protein folding if I remember correctly. Sure, it could be used for other purposes, but so could any computer. The project you may be thinking of is called ASCI White . Here's the ASCI project (Accelerated Strategic Computing Initiative).

to quote the article:

[SHEENmaster]

IBM expects the Unix offshoot to be more popular than its own version of Unix, called AIX

The truth is that AIX isn't entirely IBM's property, and Linux is not Unix. I guess SCO has an operative inside of zdnet.

Funny how Apple makes supercomputers with IBM's chips while IBM makes supercomputers with AMD's chips. Sun is starting to us x86 and Sparc64 chips despite its own UltraSparc line. HP dropped the Saturn chip for ARM. Can anyone afford their own chips these days?

Re:to quote the article:

[cabbey]

IBM makes supercomputers with AMD's chips.

Where? BuleGene uses a varient of PPC970 chips from IBM.