Re:AI chip company SambaNova Systems raises $56M
by throwaway84742 on Thursday March 15, 2018 @09:11PM
(#16597862)
Movidius kind of did, as a part of Intel. They sold dozens of their sticks, maybe even hundreds. By EOY “good enough” built in power efficient acceleration will be available on sub-$100 dev boards. So they have to raise while they still can.
Re:AI chip company SambaNova Systems raises $56M
by godelmachine on Friday March 16, 2018 @04:57AM
(#16599424)
Please correct me if I am wrong, but Movidius also supplied Vision Processing Unit to Tesla. They terminated their contract with Tesla over some tussle which I'm not aware of.
Re:AI chip company SambaNova Systems raises $56M
by throwaway84742 on Friday March 16, 2018 @12:12AM
(#16598567)
There’s no real stack without an NDA, if you want to write code for the device itself. Clang does support the SHAVE cores they use, but I don’t know what the tooling actually looks like. Probably much like any other cross compilation tool chain.
If you just want to use it, there’s a Python SDK which you can use to run a subset of models popular frameworks can train. In that case it’s pretty easy to get going. I believe there’s also a C++ integration of some sort, but I haven’t used it myself.
It’s all moot, though, because properly programmed, recent ARM CPUs are fast enough to run usable models such as MobileNet and ShuffleNet with good performance. The problem is that at the moment this draws anywhere between 4-12W if things are running full blast, at least until throttling kicks in. Movidius is more economical in terms of joules per inference, but it also costs quite a bit, and it’s not built in, so market penetration is non-existent.
As I mentioned, though, this should start being less of an issue with the advent of cheap, built in tensor acceleration IP.
Re:AI chip company SambaNova Systems raises $56M
by throwaway84742 on Friday March 16, 2018 @12:18AM
(#16598596)
You should take a look at their architecture diagram. There’s a lot of meat on the bones there besides DSP, particularly when you look at their memory setup and consider how much bandwidth it can provide simultaneously. And let us not forget that convolution is essentially a DSP operation.
I was, frankly, surprised by the technical depth of what they were able to put together.
Re:AI chip company SambaNova Systems raises $56M
by lnsru on Friday March 16, 2018 @05:20AM
(#16599488)
I have no experience with AI, but is this application very different from ltc or eth fpga mining? Do I get it right, that AI chip needs plenty of memory to hold coefficients, manipulate data, get next set of coefficients, repeat this multiple times and the bottleneck is memory bandwidth? Or it can be parallelized with more FPGAs for different layers with corresponding storage for coefficients.
Re:AI chip company SambaNova Systems raises $56M
by memebox3v on Friday March 16, 2018 @04:10AM
(#16599291)
But its still all von neumann architecture. We need a clean break at the hardware level to really get somewhere on this imho. I think reservoir computing is the way to go, with a specially designed substrate on silicon acting as the reservoir. Thats where i will be investing if I get the chance.
If you just want to use it, there’s a Python SDK which you can use to run a subset of models popular frameworks can train. In that case it’s pretty easy to get going. I believe there’s also a C++ integration of some sort, but I haven’t used it myself.
It’s all moot, though, because properly programmed, recent ARM CPUs are fast enough to run usable models such as MobileNet and ShuffleNet with good performance. The problem is that at the moment this draws anywhere between 4-12W if things are running full blast, at least until throttling kicks in. Movidius is more economical in terms of joules per inference, but it also costs quite a bit, and it’s not built in, so market penetration is non-existent.
As I mentioned, though, this should start being less of an issue with the advent of cheap, built in tensor acceleration IP.
I was, frankly, surprised by the technical depth of what they were able to put together.
Of course it has high bandwidth, it's a multi bank on chip memory...
And yes, I've seen all public (and some non public) information on their architecture
https://www.movidius.com/news