CODEWITHSUNDEEP
nvidiadeep-learning-ami
Netanel
Netanel

Reputation: 519

Nvidia-smi fails on AWS's g4ad server with deep learning AMI

I'm trying to use AWS's g4ad instance with deep learning AMI (Deep Learning AMI (Ubuntu 16.04) Version 50.0) and when I'm trying to execute:

nvidia-smi

I'm getting:

NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.

Nevertheless, it seems that CUDA is working fine. When executing nvcc --version, I'm getting:

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Mon_Oct_12_20:09:46_PDT_2020
Cuda compilation tools, release 11.1, V11.1.105
Build cuda_11.1.TC455_06.29190527_0

Upvotes: 1

Views: 1869

Answers (1)

Netanel
Netanel

Reputation: 519

Apparently, g4ad instances don't have an Nvidia GPU, but an AMD GPU. With g4dn it works perfectly.

Upvotes: 3

Related Questions