Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Which GPU(s) to Get for Deep Learning

timdettmers.com

Post reply on HN
Print view View original

11–20 of 119 posts

Re: Which GPU(s) to Get for Deep Learning

#11

Post by antognini »

ericfrederich wrote: post #7

I'm guessing you don't... you use the cloud, right?

For a hobbyist, it probably makes sense to just use some GPU instances on AWS. But it doesn't really take that long for it to become cost-effective to buy a GPU. It only takes a couple of months of full-time use before it becomes cost-effective to purchase the GPU. Even if you're building the entire machine from scratch it takes under a year. (At least, that was my estimate when I was building a rig earlier this year.) If you're doing any serious deep learning projects it's pretty easy to have a model training nearly 100% of the time.

Re: Which GPU(s) to Get for Deep Learning

#12

Post by chronic940 »

sgt101 wrote: post #6

No mention of k80's k40's...

K80s and K40s have a K. That's two generations old. We are currently on P and about to be V. If you think K80s are good, you are far far behind the times in machine learning. A single 1080Ti even in a 4U server outperforms 2 K80s.

Re: Which GPU(s) to Get for Deep Learning

#14

Post by sp332 »

AlphaSite wrote: post #13

Interestingly, AMD is claiming that their upcoming Vega Frontier Editions chips outperform Nvidia's current P100 Deep Learning chip. http://hexus.net/media/uploaded/2017/5/30f5633b-1bbf-49b7-9f...

I think signs are pointing to Vega being a HPC beast. But what they really need is the software ecosystem and support, and so far that hasn't been there. So while the new hardware looks cool, I'm really waiting for an announcement that OpenCL tooling got a lot better, or that CUDA is getting first-party support from AMD, anything to tempt those customers away from nvidia.

Re: Which GPU(s) to Get for Deep Learning

#15

Post by nightski »

antognini wrote: post #11
ericfrederich wrote: post #7

I'm guessing you don't... you use the cloud, right?

For a hobbyist, it probably makes sense to just use some GPU instances on AWS. But it doesn't really take that long for it to become cost-effective to buy a GPU. It only takes a couple of months of full-time use before it becomes cost-effective to purchase the GPU. Even if you're building the entire machine from scratch it takes under a year. (At least, that was my estimate when I was building a rig earlier this year…

Not to mention, I can imagine it's frustrating uploading many gigs of data to amazon.

Re: Which GPU(s) to Get for Deep Learning

#16

Post by sgt101 »

chronic940 wrote: post #12
sgt101 wrote: post #6

No mention of k80's k40's...

K80s and K40s have a K. That's two generations old. We are currently on P and about to be V. If you think K80s are good, you are far far behind the times in machine learning. A single 1080Ti even in a 4U server outperforms 2 K80s.

Ok - p100's? The thing is that a K80 costs.... £4k what is it that people are paying for?

Re: Which GPU(s) to Get for Deep Learning

#17

Post by treehau5 »

nightski wrote: post #15
antognini wrote: post #11

Earlier quoted context omitted.

For a hobbyist, it probably makes sense to just use some GPU instances on AWS. But it doesn't really take that long for it to become cost-effective to buy a GPU. It only takes a couple of months of full-time use before it becomes cost-effective to purchase the GPU. Even if you're building the entire machine from scratch it takes under a year. (At least, that was my estimate when I was building a rig earlier this year…

Not to mention, I can imagine it's frustrating uploading many gigs of data to amazon.

extremely frustrating. Internet service is getting worse this in country, not better, sadly. The only lucky ones are those who live in competitive markets (lol) and those who had google fiber break some ground.

The anti-competitiveness is getting worse too. Charter just took over TWC in my area.

Re: Which GPU(s) to Get for Deep Learning

#18

Post by mattnewton »

paulsutter wrote: post #5
Analemma_ wrote: post #4

Earlier quoted context omitted.

The Titan Xp is better, but it's not so much better that it's worth almost twice the price.

Two 1080TIs are way better than one Titan Xp and cost about the same (2x$700 vs $1200). Each TI has 11GB of RAM vs the TitanX's 12GB, and each TI is nearly as fast. At the NVIDIA conference all the second tier hosting companies were promoting the P100 (at NVIDIAs insistence) but when pressured admitted that their big customers now deploy 1080TIs. Paying for P100s is sort of a clown move even if you're spending someon…

Usually, but the caveat is they are only better for workloads that are easily paralizable (like hyperparameter search). Multi-gpu models are still very complicated in my experience. And the Xp is a bit faster and has 1gb more memory, which is useful in some edge cases (like large or 3D covnets)

Also if there do build with multiple gpus, watch out that you can give both cards a full 16 pci lanes, not a given in a lot of motherboards.

Re: Which GPU(s) to Get for Deep Learning

#19

Post by dragandj »

sp332 wrote: post #14
AlphaSite wrote: post #13

Interestingly, AMD is claiming that their upcoming Vega Frontier Editions chips outperform Nvidia's current P100 Deep Learning chip. http://hexus.net/media/uploaded/2017/5/30f5633b-1bbf-49b7-9f...

I think signs are pointing to Vega being a HPC beast. But what they really need is the software ecosystem and support, and so far that hasn't been there. So while the new hardware looks cool, I'm really waiting for an announcement that OpenCL tooling got a lot better, or that CUDA is getting first-party support from AMD, anything to tempt those customers away from nvidia.

CUDA is getting support from AMD in the form of HIP. In my opinion, the problem is different: even if AMD supported CUDA proper, that still won't lead them anywhere because the reason people use Nvidia is not so much CUDA itself (although it has good tools) but Nvidia's proprietary libraries: cuBLAS, cuDNN, cuFFT etc.

If AMD provided good implementations of similar libraries with OpenCL and shipped them with their drivers (similarily to Nvidia's CUDA toolkit) that would be much better than supporting CUDA, but leaving the ecosystem bare bones again...

Post reply on HN