Google supercharges machine learning tasks with TPU custom chip • Hacker News Board

Re: Google supercharges machine learning tasks with TPU custom chip

#41

Post by azinman2 » Wed, May 18, 2016, 7:52 PM UTC

fiatmoney wrote: post #27

azinman2 wrote: post #21

Earlier quoted context omitted.

Ndvidia has to be general purpose. This is not and thus can be better optimized.

"General purpose" isn't that general, if you look at the actual operations they support and their threading model. It's already fairly optimized for these sorts of operations, and this amount of claimed headroom makes me suspicious.

The fact that I can compile arbitrary programs for the GPGPU means it is general purpose. NVIDIA isn't writing softmax or backprop into silicon as a CPU instruction.

Look at how much faster ASICs for bitcoin mining are than the GPU... orders of magnitude.

Re: Google supercharges machine learning tasks with TPU custom chip

#42

Post by nickysielicki » Wed, May 18, 2016, 7:53 PM UTC

honkhonkpants wrote: post #22

nickysielicki wrote: post #8

Earlier quoted context omitted.

There's no way Google lets this leave their datacenters. Chip fabrication is a race to the bottom at this point. [1] Google is doubling down on hosting as a source of future revenue, and they're doing that by building an ecosystem around Tensorflow. What I think is interesting is how weak Apple looks. Amazon has the talent and money to be able to compete with Google on this playing field. Microsoft is late, but they…

Apple designs their own CPUs. I think they'd be able to field a massively parallel FMAC chip if they thought that was a good idea. Where Apple really looks weak is in datacenters, networking, and cloud services.

What does the iPhone of 2021 look like?

I get the feeling from today's announcements that Google sees the 2021 version of Google Now as the selling point for their 2021 Nexus line.

I don't think Apple is preparing to compete on that.

Re: Google supercharges machine learning tasks with TPU custom chip

#43

Post by Retric » Wed, May 18, 2016, 7:55 PM UTC

dgacmu wrote: post #32

eva1984 wrote: post #23

Earlier quoted context omitted.

Meaning even deeper level of vendor lock -- now you cannot even find the chips to run your application elsewhere!

No - tensorflow is open source and you can run it on many platforms. TPUs are about efficiency. You might not be able to do image recognition as efficiently without one, but you can still perform exactly the same tasks. (I work on TF this year.)

I would be shocked if tensorflow optimizations where useful 1:1 for stock Intel chips or GPU's. So, there is still plenty of lock-in even if your process runs. GPU vendors love to play this game by helping optimize games.

Re: Google supercharges machine learning tasks with TPU custom chip

#44

Post by cven714 » Wed, May 18, 2016, 7:56 PM UTC

saganus wrote: post #36

Is that a Go board stick to the side of the rack? Maybe they play one move every time someone gets to go there to fix something? or could it be just a way of numbering the racks or something eccentric like that?

Likely it was one of the racks used by AlphaGo

Re: Google supercharges machine learning tasks with TPU custom chip

#45

Post by triangleman » Wed, May 18, 2016, 7:57 PM UTC

saganus wrote: post #36

Is that a Go board stick to the side of the rack? Maybe they play one move every time someone gets to go there to fix something? or could it be just a way of numbering the racks or something eccentric like that?

It appears to be a commemorative plaque from when they defeated world champion Lee Sodol.

Re: Google supercharges machine learning tasks with TPU custom chip

#46

Post by bgalbraith » Wed, May 18, 2016, 7:57 PM UTC

harigov wrote: post #6

How is this different from - say - synthetic neurons that IBM is working on, or what nvidia is building?

IBM's TrueNorth chip is taking a much more neuromorphic design approach by trying to approximate networks of biological neurons. They are investigating a new form of computer architecture away from the classic Von Neumann model.

TPUs are custom ASICs that speed up math on tensors i.e. high-dimensional matrices. Tensors feature prominently in artificial neural networks, especially the deep learning architectures. While GPUs help accelerate these operations, they are optimized first and foremost for video rendering/gaming applications -- compute-specific features are mostly tacked on. TPUs are optimized solely for doing ML-related computations.

Re: Google supercharges machine learning tasks with TPU custom chip

#47

Post by fiatmoney » Wed, May 18, 2016, 7:59 PM UTC

azinman2 wrote: post #41

fiatmoney wrote: post #27

Earlier quoted context omitted.

"General purpose" isn't that general, if you look at the actual operations they support and their threading model. It's already fairly optimized for these sorts of operations, and this amount of claimed headroom makes me suspicious.

The fact that I can compile arbitrary programs for the GPGPU means it is general purpose. NVIDIA isn't writing softmax or backprop into silicon as a CPU instruction. Look at how much faster ASICs for bitcoin mining are than the GPU... orders of magnitude.

"Backprop" isn't even close to something that would be a "CPU instruction", it's an entire class of algorithm. It's like saying "calculus" should be a CPU instruction. Matrix multiplication & other operations, on the other hand, do neatly decompose into such instructions, which have been implemented by NVidia et al., since that's the core set of functionality they've been pushing for like a decade now.

Additional die space on additional functionality might hurt the power envelope (which is where the focus on performance / watt rather than performance kicks in) but it doesn't make your chips slower per se.

Re: Google supercharges machine learning tasks with TPU custom chip

#48

Post by semisight » Wed, May 18, 2016, 7:59 PM UTC

codecamper wrote: post #5

semisight wrote: post #3

This is huge. If they really do offer such a perf/watt advantage, they're serious trouble for NVIDIA. Google is one of only a handful of companies with the upfront cash to make a move like this. I hope we can at least see some white papers soon about the architecture--I wonder how programmable it is.

I don't see any mention of offering these chips for sale. You can rent them it seems via cloud offerings & that's it.

Yup, I assume they're gonna keep them in house as a competitive advantage for a time. I doubt they'll do it forever; the most valuable part of NVIDIA's CUDA is the ecosystem, and I think Google knows that.

Re: Google supercharges machine learning tasks with TPU custom chip

#49

Post by dharma1 » Wed, May 18, 2016, 8:00 PM UTC

semisight wrote: post #3

This is huge. If they really do offer such a perf/watt advantage, they're serious trouble for NVIDIA. Google is one of only a handful of companies with the upfront cash to make a move like this. I hope we can at least see some white papers soon about the architecture--I wonder how programmable it is.

it's ASIC tuned for specific calculations, I'm sure it's better power consumption than general purpose GPUs. Same as crypto mining ASIC's crush GPU's in terms of power efficiency.

There isn't much data yet but I'm also guessing they probably have access to much more RAM than NVidia cards and can process much bigger data sets

Re: Google supercharges machine learning tasks with TPU custom chip

#50

Post by semisight » Wed, May 18, 2016, 8:01 PM UTC

nickysielicki wrote: post #8

semisight wrote: post #3

This is huge. If they really do offer such a perf/watt advantage, they're serious trouble for NVIDIA. Google is one of only a handful of companies with the upfront cash to make a move like this. I hope we can at least see some white papers soon about the architecture--I wonder how programmable it is.

There's no way Google lets this leave their datacenters. Chip fabrication is a race to the bottom at this point. [1] Google is doubling down on hosting as a source of future revenue, and they're doing that by building an ecosystem around Tensorflow. What I think is interesting is how weak Apple looks. Amazon has the talent and money to be able to compete with Google on this playing field. Microsoft is late, but they…

I'm not sure I agree with you, long term (about the chips). I think that the value here is in the ecosystem. If Google can compete with CUDA, they'll be doing really well.