How to find the compute compatibility of my GPU that is not listed on the official website?

Accelerated Computing CUDA CUDA Programming and Performance
1

Hi everyone,
  The GPU I am using is Tesla V100, and I read the official website but failed to find its compute compatibility. The most similar one is Nvidia V100 with compute capability 7.0, but I am unsure if they have the same compute compatibility even though they are based on the same architecture.
  So my question is how to find the compute compatibility of Tesla V100?
Any help will be appreciated.

Jack

2

The NVIDIA V100 and the Tesla V100 are the same GPU.

One method to get the compute capability in your situation (“I am using…”) is to run the deviceQuery sample code.

3

Thanks for your prompt reply.
But I can’t find any documentation that indicates that The NVIDIA V100 and the Tesla V100 are the same GPU.

Why do they have different names if they are the same things?

I hope I can get the compute compatibility without running any code on real hardware because I am not always able to have access to the hardware.

Jack

4

I think it is just a branding thing. As far as I am aware, NVIDIA retired the Tesla brand, but were still actively selling the V100 GPU at the time. Or it may be a difference between the North American and European markets, based on the two links below (first one is “en-gb”, second one is “en-us”):

1 Like
5

The documentation still uses Tesla A, Tesla T as branding name. But I think your answer is more reasonable.

Thanks.

6

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.