Skip to main content

Tutorial: Installing CUDA 8 on Ubuntu 16

Introduction​

CUDA is a parallel computing platform and programming model that makes using a GPU for general-purpose computing simple and elegant. Developers can program in familiar languages like C, C++, Fortran, and others, incorporating extensions through basic keywords.

These keywords allow developers to express massive parallelism and direct the compiler to the parts of the application that will run on the GPU.

Below are the steps for installing CUDA 8 and cuDNN 7.1.4 on Ubuntu 16, along with removing the latest NVIDIA packages on a new instance.

Installation Steps​

Perform System Update and Upgrade​

First, update and upgrade the system packages:

    apt update
	apt upgrade
	reboot

Remove the existing installed Cuda Packages

    apt remove cuda-*
	apt autoremove

Install Nvidia driver which supports Tesla v100

    wget http://us.download.nvidia.com/tesla/440.33.01/NVIDIA-Linux-x86_64-440.33.01.run
	chmod +x NVIDIA-Linux-x86_64-440.33.01.run
	./NVIDIA-Linux-x86_64-440.33.01.run

To check nvidia driver output

nvidia-smi

ignore cuda version which is seen in this output

Install Cuda8 debian packages


        wget https://developer.nvidia.com/compute/cuda/8.0/Prod2/patches/2/cuda-repo-ubuntu1604-8-0-local-cublas-performance-update_8.0.61-1_amd64-deb
        dpkg -i cuda-r	epo-ubuntu1604-8-0-local-cublas-performance-update_8.0.61-1_amd64-deb


Reboot the system

reboot

To check Cuda version

    nvcc --version

To Install cuDNN Version 7.1.4 for CUDA​

  1. Download the three required packages for Ubuntu 16 from the cuDNN archive.

  2. Follow the installation and verification steps provided in the official cuDNN installation guide for detailed instructions.

These steps will ensure that cuDNN is properly installed and ready to work with CUDA 8 for your machine learning applications.