2
votes

I am currently working as a working student and now I have trouble installing Tensorflow-gpu on a machine using a Nvidia Quadro GV100 GPU.

On the Tensorflow homepage I found out that I need to install CUDA 9.0 and Cudnn 7.x in order to run Tensorflow-gpu 1.9. The problem is that I can't find a suitable CUDA version supporting the GV100. Could it be that there is no CUDA version yet? Is it possible that one can't use the GV100 for tensoflow-gpu?

Sorry for the stupid question, I am new to installing DL frameworks :-)

Thank you very much for your help!

1
Where are you seeing the supported/unsupported information? I'm pretty sure that as long as your device looks good for CUDA (this random page for instance lists CUDA support: pny.com/nvidia-quadro-gv100).Svend
@Svend Yes, I found that information too. But having a look here link, the GV100 is not listed as supported GPU. Additionally, I already tried to download CUDA 9.0 and to install it. The first step of the installation is an automatic system check for compatibility of your GPU - this results in a kind of "no compatible hardware was found" error :-( But can it really be true that their top-knotch-gpu is not supported?Midas.Inc
It's perhaps the issue that you're a bit too advanced maybe? Just from googling around, I can't imagine it doesn't support CUDA. This whitepaper images.nvidia.com/content/volta-architecture/pdf/… mentions GV100 as having the highest Compute Cabability level (7.0). The marketing speak talks about "cuda" and "tensor" cores, if that's not CUDA enabled, I'd be shocked. Talk to whoever sold this card to you, or just reach out to NVIDIA directly. You might try to reach out directly to one of NVIDIAs twitter accounts. There must be a download somewhere :)Svend
Yeah, I thought exactly the same. I would have been shocked, too :D Thanks again for your help!Midas.Inc

1 Answers

4
votes

On the Tensorflow homepage I found out that I need to install CUDA 9.0 and Cudnn 7.x in order to run Tensorflow-gpu 1.9.

That is if you want to install a pre-built Tensorflow binary distribution. In that case you need to use the version of CUDA which the Tensorflow binaries were built against, which in this case in CUDA 9.0

The problem is that I can't find a suitable CUDA version supporting the GV100

The CUDA 9.0 and later toolkits fully support Volta cards and that should include the Quadro GV100. The driver which ships with CUDA 9.0 is a 384 series which won't support your GPU. If you are referring to a driver support issue, then the solution would be to install the recommended driver for your GPU, and only install the CUDA toolkit from the CUDA 9.0 bundle, not the toolkit and driver, which is the default.

Otherwise you can use CUDA 9.1 or 9.2, which should have support for your GPU with their supplied drivers, but you will then need to build Tensorflow yourself from source.