Slashdot Mirror

← Back to Stories (view on slashdot.org)

$30M Stampede 2 Supercomputer To Provide 18 Petaflops of Power To Researchers Nationwide (techcrunch.com)

Posted by BeauHD on from the free-for-all dept.

An anonymous reader writes: Funded by grants from the National Science Foundation and built at the University of Texas at Austin, the Stampede 2 supercomputer looks to contend with the global supercomputer Top 5. With 18 petaflops of processing power, it aims to help any researcher with a problem requiring intense number crunching. For example, atomic and atmospheric science simulations would take years to work-out on a desktop PC but only days on a supercomputer. Texas Advanced Computing Center director Dan Stanzione said in a UT press release, "Stampede has been used for everything from determining earthquake risks to help set building codes for homes and commercial buildings, to computing the largest mathematical proof ever constructed." The Stampede 2 is about twice as powerful as the original Stampede, which was activated in March of 2013. Instead of the 22nm fabrication tech in the original Stampede, the Stampede 2 will feature 14nm Xeon Phi chips codenamed "Knights Landing" forming 72 cores compared the original system's 61 cores. With double the RAM, storage and data bandwidth, the Stampede 2 can shift up to 100 gigabits per second, and its DDR4 RAM can perform fast enough to work as a third-level cache as well as fulfill ordinary memory roles. In addition, it will feature 3D Xpoint non-volatile memory. It will be at least a year before the Stampede 2 is powered up since it just received funding.

44 comments

  1. who needs super computers nowadays by Anonymous Coward · · Score: 0

    .. when we can do everything in the cloud?
    My man at IBM says Watson will be able to do everything due time.

  2. This is a fucking press release by ChesterRafoon · · Score: 2

    "It will be at least a year before the Stampede 2 is powered up since it just received funding" That's some real hard news there Slashdot. Thanks.

    1. Re:This is a fucking press release by Anonymous Coward · · Score: 0

      "It will be at least a year before the Stampede 2 is powered up since it just received funding"

      That's some real hard news there Slashdot. Thanks.

      Especially since it will contains chips we don't sell yet. I don't know where the hell they got a datasheet from.

    2. Re:This is a fucking press release by serviscope_minor · · Score: 1

      FFS. News about upcoming supercomputers IS news for nerds. See the thing about the xpoint memory too?

      --
      SJW n. One who posts facts.
    3. Re:This is a fucking press release by Guybrush_T · · Score: 1

      The way it is written also looks like a press release written for the average non-nerd. "Number crunching" ? No, climate simulation. "Years on a desktop-PC" ? yeah, we can do the math. And desktop PC doesn't really make any sense (I can fit a number of GPUs on my desktop PC)

  3. Finally... by Anonymous Coward · · Score: 0

    a machine you can run eclipse on.

  4. How many Apollo 11 computers? by Anonymous Coward · · Score: 0

    Can it take us to Mars, the Bringer of Evil?

    1. Re:How many Apollo 11 computers? by Anonymous Coward · · Score: 3, Interesting

      This page claims:

      The Apollo Guidance Computer weighed 70 pounds, consumed 55 watts of power and occupied only 0.97 cubic feet inside the spacecraft. These first digital flight computers were limited to only 36,000 words of fixed memory and 2,000 words of RAM, and operated at a 12-microsecond clock speed.

      That was 16 bit words. 12 microseconds is about 83K-ops/sec. Unclear how many cycles each operation took, but let's give it the benefit of the doubt and say the throughput was instruction per clock. (Probably optimistic).

      Those 18-petaflops are probably 32 bit, if not 64 or 80. Let's say each is worth two of the AGC 16-bit ops.

      So, 18 petaflops would be similar to 4.3 x 10^11 apollo guidance computers. You can probably swing that by 100X in either direction depending on the assumptions you make, but let's roll with it.

      Each AGC weighed 32 kg. 4x10^11 of them would weigh in at ~ 1.4x10^13 kg. For comparison, this is roughly 35 times the combined mass of the entire human population.

  5. Re:problems requring intense number crunching by Anonymous Coward · · Score: 0

    or pyOffice

  6. Imagine... by MAXOMENOS · · Score: 0

    ...a Beowulf cluster of....

    (Dodges rotten fruit)

    --
    Finding God in a Dog
    1. Re:Imagine... by Hognoxious · · Score: 1

      If half of the AI hype was true the Beowulf cluster would imagine YOU!!!

      --
      Confucius say, "Find worm in apple - bad. Find half a worm - worse."
  7. Re:These resources will be wasted by Anonymous Coward · · Score: 0

    Economics is nothing but fiction too, but I don't see you complaining about the wasted resources there.

  8. Moore's law dead? by Enigma2175 · · Score: 3, Insightful

    So this new computer is twice as powerful as the original Stampede, which cost $27 million and came online in 2013. It sounds like Stampede 2 will come online in 2018 at a cost of $30 million. So over 5 years the power only doubled for the same cost - if computing power per cost unit doubles every 18 months like Moore predicted, why isn't Stampede 2 like 16 times more powerful than its predecessor?

    --

    Enigma

    1. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      Moore said density (transistor count), not your "power". Must I always correct those who failed History? Change yourstory as needed to fit the crime. That won't change breaking The Law.

    2. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      Moore's Law cannot compete with the iron axiom of minimization--of expenses incurred relative to government grant money received.

    3. Re:Moore's law dead? by byrtolet · · Score: 3, Interesting

      Moore said density (transistor count), not your "power". Must I always correct those who failed History?

      No, you must not!

      Moore's law is not a law, it's just an observation (and maybe self-fulfilling prophecy). And the observation of computers getting more and more powerful was also correct. So why not associate one with the other!

    4. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      1. That's not what Moore's law said. (read all the other posts, referencing you as an idiot).
      2. Moore's law be dammed, the thing still won't run Crysis.

    5. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      No, Its not a computer, its a comuter cluster Moores law is long dead..

    6. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      Perf diff of last four i7 from intel is about what, 15% maybe? Back four is sky (2015), broad, has, ivy (2012), or three years (36 mo). Look at transistor count and package size. Yes, dammit, cache counts. Do the mathz!

    7. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      Moore's law is not a law, it's just an observation (and maybe self-fulfilling prophecy).

      More of an economic law, at this point, to whatever company can't keep up.

    8. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      What? It's not a law? You mean, I won't go to jail if I break it?
      That's really good news. Now I won't have to worry about being sent to jail and have my anus penetrated by a criminal's erect penis!

    9. Re:Moore's law dead? by Victotronics · · Score: 1

      That is a fair comment. The multiplication factor is somewhat low for the amount of time elapsed. However, Stampede-1 had a peak performance that was somewhat hard to reach for any application than the Linpack benchmark. That is because of the Knight's Corner processor. The new Knight's Landing in Stampede-2 is considerably easier to use, making a larger fraction of peak effectively usable.

    10. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      So this new computer is twice as powerful as the original Stampede, which cost $27 million and came online in 2013. It sounds like Stampede 2 will come online in 2018 at a cost of $30 million. So over 5 years the power only doubled for the same cost - if computing power per cost unit doubles every 18 months like Moore predicted, why isn't Stampede 2 like 16 times more powerful than its predecessor?

      Supercomputer performance measures include more than CPU clock speeds.

    11. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      Moore's law is that transistor counts double every 18 months, not processing speed or power. And the 18 month thing is slowing.

    12. Re:Moore's law dead? by Anonymous Coward · · Score: 0

      Moores laws is just something some guy said which marketing people latched on to because it allows for them to sell new stuff at a predictable rate which meshes well with what people are willing to pay. At several points in the past couple of decades it has either proven way too high or too low of an estimation.

    13. Re:Moore's law dead? by oldcarsmell · · Score: 1

      Transistor count != FLOPS

  9. Re:problems requring intense number crunching by Anonymous Coward · · Score: 0

    Is this thing available for hire? Because I, for one, am still trying to boot Vista...

    That's what you get for upgrading from XP. XP booted pretty fast.

  10. Bitcoin? by Anonymous Coward · · Score: 0

    But how fast can it mine Bitcoin?

    1. Re:Bitcoin? by ShanghaiBill · · Score: 2

      But how fast can it mine Bitcoin?

      It is not an appropriate architecture for bitcoin mining. Much of the design is focused on moving massive amounts of data in, out, and between the cores. You don't need any of that for bitcoin, since each hash is independent and uses very little data. Bitcoin mining can be done way more cost effectively on ASICs.

  11. Introducing another Vapor Super Computer by Anonymous Coward · · Score: 1

    Since TFA is about a Super Computer that has yet to be built, please allow me to introduce two links to another Super Computer that has yet to be built

    http://www.nextplatform.com/20...

    http://www.nextplatform.com/20...

  12. Moore's law dead? We Are Doomed! by Anonymous Coward · · Score: 0

    Doomed, I say. Why can't everything keep growing all the time? Exponentially, at that?

    Aaaaah!

  13. DDR4 as L3 cache, really? by fuzzyfuzzyfungus · · Score: 1

    I realize that DDR4 is pretty fast; and given how relatively cheap it is it is certainly good for what you pay; but is it honestly anywhere close to being comparable to L3 cache? I would think that, even if you spared absolutely no expense and went with the fastest RAM money could buy, signal propagation delays would, at contemporary clock speeds, make system RAM tens of cycles slower to access than anything on die.

    1. Re:DDR4 as L3 cache, really? by Anonymous Coward · · Score: 0

      It's summarily nonsense. The six channels of DDR4 is used as working memory, with the preference of using the 16 GB of HMC, or MCDRAM as Intel likes to call it. The HMC memory is fast enough to be used as cache for that older software.

  14. There has got to be a HURD joke in here somewere by Anonymous Coward · · Score: 0

    n/t

  15. I often wonder by TheCastro1689 · · Score: 1

    How much computing power is wasted/used up by crappily designed programs that anticipate having lots of RAM access?

    1. Re:I often wonder by Arkh89 · · Score: 1

      You have no idea how bad scientific code is : bad memory accesses, poor algorithms designs, dumb data structures, no consideration of possible bottlenecks, no cluster architecture knowledge...
      And wait until you see people allocating either way too much or far less resources than required for their jobs in these clusters...

    2. Re:I often wonder by Anonymous Coward · · Score: 0

      Vast amounts. That's why there all these distributed applications that use "spare clock cycles" have become popular. Though, anywhere that makes use of high-performance parallel processing hardware has their own dedicated team of code optimizing staff who clean up the code before it is run on any supercomputing system.

  16. Seems underperforming... by Khyber · · Score: 1

    Only 100 gigabits can be moved per second? You can't move one bit per clock cycle or even every four?

    --
    Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
    1. Re:Seems underperforming... by oldcarsmell · · Score: 1

      That's likely the inter-node cluster network, the data transfer speed between each node in the cluster. This is not a single computer, it is a cluster of high-end computers, you know.