NOAA’s petascale computer for climate research coming together at Oak Ridge lab

gaea1.jpg
ORNL photos/Jay Nave
Jamison Daniel of ORNL did the graphics for the Gaea cabinets. In photo below, Al Enger, right, and Scott Morgan of Cray work on installation of the new Gaea components.
gaea2.jpgCray recently delivered the final 26 cabinets of the National Oceanic and Atmospheric Administration’s Gaea climate research supercomputer, which is housed at Oak Ridge National Laboratory. The newly arrived cabinets are loaded with the new AMD 16-core Interlagos processors. According to Jeff Nichols, an associate lab director at ORNL who heads the computational science directorate, the Gaea system is still in two pieces. The first piece is the original 14-cabinet system with a peak capability of 260 teraflops, Nichols said. The second piece is the new 26-cabinet system with a capability of 720 teraflops, he said.
After the first piece is upgraded in the spring and the two pieces are integrated into one system, Gaea will become a 1.1 petaflops supercomputer, ORNL’s computing chief (who returned from a visit to China last week, where he spoke at a conference) said.


Gaea is NOAA’s prime supercomputing resource, and it will become the third petascale machine housed at ORNL. Jaguar, soon to be morphed into Titan, and Kraken, a National Science Foundation machine, are the others.
According to info provided by ORNL, Gaea is liquid-cooled and uses Cray’s ECOphlex technology, which employs a refrigerant to remove most of the 2.2 MW heat load. “The technology is significantly more energy-efficient than the air-cooling systems typically found in other leading-edge HPC systems, the lab said.
The system, which is used for climate modeling and resource, also includes two separate Lustre parallel file systems “that handle data sets that rank among the world’s largest,” ORNL said. “NOAA research partners access the system remotely through speedy wide area connections. Two 10-gigabit (billion bit) lambdas, or optical waves, pass data to NOAA’s national research network through peering points at Atlanta and Chicago.”

This entry was posted in ORNL on by .

About Frank Munger

Senior Writer Frank Munger covers the Dept. of Energy's Oak Ridge facilities and many related topics — nuclear weapons, nuclear waste and other things nuclear, environmental cleanup and science of all sorts. Atomic City Underground is, first and foremost, a news blog, but there's room for analysis, opinion and random thoughts that have no place else to go.

33 thoughts on “NOAA’s petascale computer for climate research coming together at Oak Ridge lab

  1. Skeptic

    Two thoughts:
    1. No matter how fast you process garbage, your output will still be garbage.
    2. I believe they should be required to operate this system using only wind and solar. Want to guess how many operations per second it will do in the middle of a windless night?

  2. Ron Moses

    My concerns would be:
    1. Will the models they feed this thing feature a built-in assumption of man-made climate change?
    2. If any resulting data calls into question the theory of man-man climate change, will that data be suppressed or discarded as has consistently been done to date?

  3. DonM

    Navier Stokes equations describe fluid flow with changes in density and temperature, aka weather and climate. They are non-linear, with sensitive dependence on initial conditions, aka chaotic. Because of the sensitive dependence on initial conditions, distant (7 days +) future states can not be accurately predicted from past states, or if you prefer, you can do all the predictions you want, but you can’t select the right one from the wrong one.
    This is another government effort that converts money and energy into noise and falsehood.

  4. BDR

    Why continue to study climate when the science is settled? Just ask The Goracle.
    Say, where is Al these days?

  5. DocRambo

    Really afraid the data entered will be biased to begin with. Will now have the ability to hide even more incriminating emails.

  6. AlGore

    Wouldn’t all that power be better allocated figuring out cold fusion or a cure for cancer or the solution to male pattern baldness (speaking of AlGore)?

  7. Dickens Goes Metro

    I bet a $20 barometer with five indicators (Stormy, Rain, Change, Fair and Very Dry) could produce equally reliable results.
    Mother Gaea is looking pretty porky by comparison.

  8. werbaz neutron

    Will the democrat fascists program this puter in “special” ways to provide the latest “proof” that human activity is changing the earth’s long run temperature oscillation?

  9. richard40

    So will all the data used, and the algorithms used to process it, be made available to the public, so we can verify they are not biased? And if others make credible charges of bias, will this group honestly try to answer those charges? So far most climate groups have refused to do that, denying the 3 most fundamental requirements for good science, transparency, replicability, and the honest search for truth. If not, it is just a faster way to process garbage inputs to get garbage answers.

  10. Zeppo

    I recall when our met got a new super to drive their equations. a program called Hirlam they made a 4x denser matrix, which should result in 4x better accuraccy. But the result was that the weatherforcast was terrible that first year, so weather is definitely not a defined science yet. Observation is still the best tool.

  11. bflat879

    I wouldn’t let James Hansen within 1000 miles of this computer. I’d also have a list of others who needn’t be close but it could go on and on. Obviously, we do need something besides East Anglia to get working on climate change.

  12. Korla Pundit

    I see they’ve included a Germit D29-07 Dropacitor powered by two 360mV CromuLate hurbines, custom-crafted to hide the decline.

  13. Moses48

    @chuckR: Please do not confuse weather and climate. I hate it so much when meteorologists attribute bad weather to climate changes.
    @Fatbuckel: The hardest part about climate science, is that it’s hard to replicate anything since timetables are all on the hundreds of years.

  14. ffakr

    Aside from the inherent bias against any peer-reviewed and supported science that opposes the stances of their political leaders, the conservative trolls hear have once again amply proved that they have a fundamental confusion about the difference between “weather” and “climate”. This won’t be used to predict the weather in Toledo 6 rather than 5 days out.
    sad, sad, little partisan denialists.
    BTW, I’m writing this while sitting about 15′ over several clusters that are running climate simulations. I’m not some ignorant little lamb who believes a politician or preacher over my colleagues who have dedicated their life’s work and substantial intellect to the honest pursuit of truth.

  15. Dennis Farr

    If you do not think that 5-6 record high temperature records being set every summer indicates the climate is changing, then I cannot talk to you. If you do, then it is either just something we cannot do anything about, and we all die, or there is something we can do about it. It’s like “I wish they didn’t have to take off my legs, but they keep saying it’s the 4 liters of Dr. Pepper I drink every day, and that’s a load of propaganda.”

  16. Frank

    This article is about climate and the clown act going on at East Anglia is about climate research. There is no confusion among the climate skeptics posting here. Provide the data, formulas and software for peer review or quit asking for my tax dollars!

  17. Dave G

    @ffakr: This thing won’t predict climate any better than weather. It just eats taxes.
    BTW, your simulations start with flawed assumptions. I don’t even have to see them to know that. Sad, sad, little partisan kool-aid drinker.
    @Dennis Farr: What do you do when we have record cold temps this winter? How do you keep your head from exploding? I know, wrap aluminum foil around your head!

  18. moke

    ffakr,
    Has your substantial intellect yet figured out that modeling climate with solar effects abstracted away as a set of normally distributed random variables is a waste of your time and everyone else’s money?
    Sincerely,
    a non-conservative skeptic

  19. George B

    Simply amazing how much spending one can create by scaring people about CO2.
    1. We know with absolute certainty that a 2 degree rise in global temperature will not be catastrophic because during this interglacial Earth was 2 degrees warmer than now for several thousand years. We and all the other species survived it.
    2. We know a 5 degree rise won’t be catastrophic, either, because the last interglacial was 5 degrees warmer than now and virtually all species here today survived that interglacial as well.
    We will run out of fossil fuel long before we get to a 5 degree anthropogenic rise.
    It is just amazing at the amount of money this issue generates.

  20. Korla Pundit

    “my colleagues who have dedicated their life’s work and substantial intellect to the honest pursuit of truth.”
    Yes, and they have to protect that wasted life and all future funding for studying this century’s Piltdown Man by silencing skeptics, fabricating data and threatening peers who review them the wrong way. That’s not science; that’s thuggery.

  21. evanh

    It’s never been a question of survival as a species. It’s more a question of how much damage to the environment that we depend on. And, by extension, how much population reduction we’ll have to manage. Preferably none at all but the longer we procrastinate on burn fossil fuels the worse we’re making for everyone.

  22. Luther Wu

    @ Korla Pundit: You suggest “population reduction”.
    Might I suggest a high bridge for yourself?
    You would have made a good Brown Shirt.
    Tool!

  23. Chuck

    -I wonder how much CO2 is created by generating 2.2 MW? 🙂
    – I believe that CO2 is creating a warmer climate, but the most we can do is slow the change down.
    -We will not all “die” from the change. There will be winners and losers. Maybe to simulations will tell us to move to Canada.
    -The cost of this science is nothing compared to what we spend researching how to blow each other up.
    -Why does everybody have to take such extreme positions? Maybe some do not want to be taxed for breathing 🙂

  24. Dean Curtis

    When you all get tired of playing with it can you set it up so the average joe can play pong or solitaire on it so we taxpaying morons can be sure we’re getting our money’s worth? Thanks.

  25. Jefff

    Most of the commenters here can’t be distinguished from that ’60s-era computer program “Parry” that tried to simulate a paranoid schizophrenic and launched into canned diatribes about the Mafia at the least cue. These guys just repeat their diatribes about climate change conspiracies instead. But don’t worry, I’m not talking about you.
    You might not know that ORNL’s power is in fact solar powered in a very inefficient way, by one of the world’s largest hydroelectric systems, the Tennessee Valley Authority’s 29 dams. Specifically, fossil fuel plants produced 62% of TVA’s total generation in fiscal year 2005, nuclear power 28%, and hydropower 10%.
    Personally, I think it’s amazing that so much compute power can fit into a mere 40 cabinets. The datacenters used by Google or Facebook hold thousands of cabinets each.

  26. kpp

    The IPCC refuses to adhere to several very basic tenants for scientific work.
    1) Allow _independant_ verification of one’s results
    2) Public the raw data, and document the methodology and analysis methods used.
    Without either, its junk science.

  27. Rus Toxicodendron

    It’s totally amazing how much all the comments add to the original info. And how enlightening most of them are.

  28. Rus Toxicodendron

    It’s totally amazing how much all the comments add to the original info. And how enlightening most of them are.

Comments are closed.