Slashdot Mirror

← Back to Stories (view on slashdot.org)

Google's Tensor Processing Unit Could Advance Moore's Law 7 Years Into The Future (pcworld.com)

Posted by BeauHD on from the Moore's-Law dept.

An anonymous reader writes from a report via PCWorld: Google says its Tensor Processing Unit (TPU) advances machine learning capability by a factor of three generations. "TPUs deliver an order of magnitude higher performance per watt than all commercially available GPUs and FPGA," said Google CEO Sundar Pichai during the company's I/O developer conference on Wednesday. The chips powered the AlphaGo computer that beat Lee Sedol, world champion of the game called Go. "We've been running TPUs inside our data centers for more than a year, and have found them to deliver an order of magnitude better-optimized performance per watt for machine learning. This is roughly equivalent to fast-forwarding technology about seven years into the future (three generations of Moore's Law)," said Google's blog post. "TPU is tailored to machine learning applications, allowing the chip to be more tolerant of reduced computational precision, which means it requires fewer transistors per operation. Because of this, we can squeeze more operations per second into the silicon, use more sophisticated and powerful machine learning models, and apply these models more quickly, so users get more intelligent results more rapidly." The chip is called the Tensor Processing Unit because it underpins TensorFlow, the software engine that powers its deep learning services under an open-source license.

11 of 86 comments (clear)

  1. Probably not advancing Moore's law by starless · · Score: 4, Insightful

    Moore's law relates to the number of components in an integrated circuit.
    I really doubt these things put more transistors onto a piece of silicon.

    1. Re:Probably not advancing Moore's law by AchilleTalon · · Score: 3, Insightful

      You are right. This is bullshit. He didn't mention anything about the density of transistors. So, since this is a specialized chip, the performance claim cannot be compared to a general purpose chip.

      --
      Achille Talon
      Hop!
    2. Re:Probably not advancing Moore's law by dbIII · · Score: 3, Insightful
      Also from the summary:

      "which means it requires fewer transistors per operation"

      So as you suggest it's definitely a misuse of the term used to roughly describe part of Intel's roadmap when Moore was there.

  2. Yeah, like DSPs... by RyanFenton · · Score: 5, Interesting

    Specialized processing chips have been several 'generations' ahead in terms of processing per dollar for many decades. In the 90's at least, DSPs were doing audio/video processing much cheaper by performing many machine-level steps simultaneously in one 'cycle' with less power than a general processor, by leaving out the features and cost of a general processor. And all you had to do to use them was test them on a hardware emulator, flash them, then pop them into production test run until you were good enough to deploy. Depending on the chip, they could run on a trickle of power, without active cooling, and match a much most costly general chip for pennies.

    I mean, it's how we got cell phones, and LOTS of other things, including most things in a computer that aren't the CPU.

    But isn't Moore's law more about transistors per unit cost, rather than performance per cost? Seems like a fundamental misunderstanding in the headline... which seems about as common as specialized chips in modern technology.

    Ryan Fenton

    1. Re:Yeah, like DSPs... by RyanFenton · · Score: 3, Informative

      Linky

      "The complexity for minimum component costs has increased at a rate of roughly a factor of two per year. Certainly over the short term this rate can be expected to continue, if not to increase. Over the longer term, the rate of increase is a bit more uncertain, although there is no reason to believe it will not remain nearly constant for at least 10 years."

      - Gordon E. Moore, April 19, 1965

      It's both cost and density, and continued to be so as it crystallized into the transistor doubling every 18 months figure. To double the density, only at exorbitant cost would not really be an increase in accessible technology. It's not just the technology being invented and monopolized, but being rolled out and usable by the entire field. Increasing computational complexity is still the most important part - but cost has always been a part of the idea too.

    2. Re:Yeah, like DSPs... by konohitowa · · Score: 3, Funny

      Perhaps is time for squirrels law, no?

  3. Re:Too superior for humanity by Anonymous Coward · · Score: 3, Insightful

    It's easy to disparage the efforts of somebody trying big things when they don't go as planned isn't it. But they are trying and the next step they take could be that 'big thing'. What have you done lately?

  4. Tensor Processing Units not new by Shinobi · · Score: 4, Informative

    Tensor Processing Units are not new. SGI used to offer that for their Octane, aimed pretty heavily at the satellite image analysis crowd.

    1. Re:Tensor Processing Units not new by Namarrgon · · Score: 3, Informative

      Right, this isn't a general-purpose DSP but a custom ASIC designed to run their TensorFlow graphs efficiently.

      --
      Why would anyone engrave "Elbereth"?
  5. Re:Google! YEY! by Anonymous Coward · · Score: 3, Informative

    Google I/O is happening at the moment.

  6. TPU sounds familiar by FreshnFurter · · Score: 3, Interesting

    Was that not something introduces about 20 years ago by Silicon Graphics? Or am I getting old
    http://manx.classiccmp.org/mir...