31–40 of 46 posts
Having done a technical review of what Nervana has to offer... Why Intel? My assumption, is that Intel will lease the chips + software, similar to the way HP, Dell, etc. Lease GPU machines. Nervana honestly woukd have been obsolete and dead pretty soon without this. I see this as an acuire-hire.
>> an acquire-hire. For $350 million is not an acquire-hire.
Update: An earlier version of the story indicated the purchase price was more than $350 million, according to a source. Multiple investors told Recode the purchase price was significantly above that price, with one pegging it at $408 million.
Earlier quoted context omitted.
> Intel probably bought 'em to work on integrated GPUs or Xeon Phi or something. Ha, ha. Yeah… no. > there's no mention of its power consumption Can't be tremendously higher than Pascal for reasons of physics. > sure, you can get a million TFlops without cache, now try to get that chip to do anything useful "Without cache" is certainly an exaggeration. It won't have a globally coherent cache hierarchy in the style of…
> > Intel probably bought 'em to work on integrated GPUs or Xeon Phi or something. > Ha, ha. Yeah… no. Oh? You don't think they'd acquihire a machine learning startup to work on their compute/machine learning offering in Xeon Phi? I would be very, very surprised if Intel introduces a new machine-learning-oriented processor—though that would be quite interesting, so let's hope you're right. > Can't be tremendously hig…
I really don't think CUDA support is important for nervana's offering. They think of themselves as the Apples of deep learning - they want to offer an integrated stack from the chip all the way to the APIs. The way most people use deep learning you don't really need to know CUDA, you just need to use a library that is fast. So it's enough that nervana's engineers know how to write deep learning libs for their own chip. Furthermore I can't see Intel caring that much about CUDA support, since CUDA is owned by Nvidia.
What I've heard about the chip makes it sound really exciting. Many of the trade-offs in deep learning are different from the ones you do in graphics, so specialized hardware makes sense.
I figured Nervana was mostly dead if they were stuck at 28 nm I figured Intel was down and out in Santa Clara without a strong deep learning play That all changed today. Intel has been bouncing all over the place trying to break into deep learning. If they don't screw this up, they just found their way in IMO.
>I figured Nervana was mostly dead if they were stuck at 28 nm Why? Is there a competitor that has leveraged a more modern process technology?
However, getting access to Intel's fabs makes them a lot more interesting and competitive. It's not a slamdunk for Intel yet because they still have to incorporate this into their product line (anyone seen Altera's Stratix 10 yet? Because that was supposed to be 2014's 10+ TFLOP GPU killer), but it's a fantastic acquisition and I wish them the best.
Earlier quoted context omitted.
> Intel probably bought 'em to work on integrated GPUs or Xeon Phi or something. Ha, ha. Yeah… no. > there's no mention of its power consumption Can't be tremendously higher than Pascal for reasons of physics. > sure, you can get a million TFlops without cache, now try to get that chip to do anything useful "Without cache" is certainly an exaggeration. It won't have a globally coherent cache hierarchy in the style of…
> You're just being silly now. CUDA isn't a standard, it's proprietary to NVIDIA I would love to agree with you but the reality is that CUDA is the de-facto industry standard e.g. [1]. > and this isn't a general purpose processor anyway. The "GP" part in GPGPU would like to disagree... [2] [1] https://en.wikipedia.org/wiki/Comparison_of_deep_learning_so... [2] http://www.nvidia.com/object/cuda_home_new.html
Case in point, GPUs are standard (for now) for training DNNs, but the big server farms that actually do the scoring run on Xeons.
Earlier quoted context omitted.
> Intel probably bought 'em to work on integrated GPUs or Xeon Phi or something. Ha, ha. Yeah… no. > there's no mention of its power consumption Can't be tremendously higher than Pascal for reasons of physics. > sure, you can get a million TFlops without cache, now try to get that chip to do anything useful "Without cache" is certainly an exaggeration. It won't have a globally coherent cache hierarchy in the style of…
> You're just being silly now. CUDA isn't a standard, it's proprietary to NVIDIA I would love to agree with you but the reality is that CUDA is the de-facto industry standard e.g. [1]. > and this isn't a general purpose processor anyway. The "GP" part in GPGPU would like to disagree... [2] [1] https://en.wikipedia.org/wiki/Comparison_of_deep_learning_so... [2] http://www.nvidia.com/object/cuda_home_new.html
Earlier quoted context omitted.
> Intel probably bought 'em to work on integrated GPUs or Xeon Phi or something. Ha, ha. Yeah… no. > there's no mention of its power consumption Can't be tremendously higher than Pascal for reasons of physics. > sure, you can get a million TFlops without cache, now try to get that chip to do anything useful "Without cache" is certainly an exaggeration. It won't have a globally coherent cache hierarchy in the style of…
> You're just being silly now. CUDA isn't a standard, it's proprietary to NVIDIA I would love to agree with you but the reality is that CUDA is the de-facto industry standard e.g. [1]. > and this isn't a general purpose processor anyway. The "GP" part in GPGPU would like to disagree... [2] [1] https://en.wikipedia.org/wiki/Comparison_of_deep_learning_so... [2] http://www.nvidia.com/object/cuda_home_new.html
Earlier quoted context omitted.
> Intel probably bought 'em to work on integrated GPUs or Xeon Phi or something. Ha, ha. Yeah… no. > there's no mention of its power consumption Can't be tremendously higher than Pascal for reasons of physics. > sure, you can get a million TFlops without cache, now try to get that chip to do anything useful "Without cache" is certainly an exaggeration. It won't have a globally coherent cache hierarchy in the style of…
> > Intel probably bought 'em to work on integrated GPUs or Xeon Phi or something. > Ha, ha. Yeah… no. Oh? You don't think they'd acquihire a machine learning startup to work on their compute/machine learning offering in Xeon Phi? I would be very, very surprised if Intel introduces a new machine-learning-oriented processor—though that would be quite interesting, so let's hope you're right. > Can't be tremendously hig…
If Intel can ape Nvidia at deep learning then it is a big growth opportunity for them.
Earlier quoted context omitted.
> You're just being silly now. CUDA isn't a standard, it's proprietary to NVIDIA I would love to agree with you but the reality is that CUDA is the de-facto industry standard e.g. [1]. > and this isn't a general purpose processor anyway. The "GP" part in GPGPU would like to disagree... [2] [1] https://en.wikipedia.org/wiki/Comparison_of_deep_learning_so... [2] http://www.nvidia.com/object/cuda_home_new.html
What happened to OpenCL? https://en.wikipedia.org/wiki/OpenCL