Skip to content

Latest commit

 

History

History
149 lines (111 loc) · 3.88 KB

install_ubuntu_for_gpu_server.md

File metadata and controls

149 lines (111 loc) · 3.88 KB
Raw

Preparing to install Ubuntu.

  • Select nothing. Just click Continue.
  • Click Continue in UEFI mode.

Install type.

  • Select Something else, which means you can create or resize partitions yourself, or choose multiple partitions for Ubuntu.

Select free space. Click the + button. And set the following.

  • Size: 1MB
  • Type for the new partition: Primary
  • Location for the new partition: Beggining of this space
  • Uses as: Reserved BIOS boot area

Select free space. Click the + button. And set the following.

  • Size: 512MB
  • Type for the new partition: Primary
  • Location for the new partition: Beggining of this space
  • Uses as: Ext4 journaling file system
  • Mount point: /boot

Select free space. Click the + button. And set the following.

  • Size: 65536MB (=64GB=RAM memory size)
  • Type for the new partition: Primary
  • Location for the new partition: Beggining of this space
  • Uses as: swap area

Select free space. Click the + button. And set the following.

  • Size: the size for the working space
  • Type for the new partition: Primary
  • Location for the new partition: Beggining of this space
  • Uses as: Ext4 journaling file system
  • Mount point: /

Then, click Install Now.

Keep choosing Continue.

Set your user name and password.

Then, you can start install Ubuntu.

Set up network

sudo apt update
sudo apt upgrade
sudo apt install build-essential cmake pkg-config libjpeg8-dev libtiff5-dev libjasper-dev libpng12-dev libgtk-3-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libatlas-base-dev gfortran screen
sudo apt-get install python2.7 python-pip python3-pip python-dev python-virtualenv git dkms linux-headers-generic openssh-server vim build-essential python3-tk nodejs npm nodejs-legacy
sudo pip3 install -U pip
sudo pip3 install -U matplotlib scipy numpy scikit-learn atpy astropy pandas readchar JSAnimation opencv-python jupyter jupyterlab pillow

Install driver

wget http://us.download.nvidia.com/XFree86/Linux-x86_64/375.66/NVIDIA-Linux-x86_64-375.66.run
sudo apt-get remove nvidia* && sudo apt autoremove
sudo vim /etc/modprobe.d/blacklist.conf

Then, add the followings in the end of the text.

blacklist amd76x_edac 
blacklist vga16fb
blacklist nouveau
blacklist rivafb
blacklist nvidiafb
blacklist rivatv
blacklist lbm-nouveau
options nouveau modeset=0
alias nouveau off
alias lbm-nouveau off

echo options nouveau modeset=0 | sudo tee -a /etc/modprobe.d/nouveaukms.conf
sudo update-initramfs -u
reboot

Ctrl+Alt+F1. Then, login.

sudo service lightdm stop

Install nvidia driver with installing CUDA.

Install CUDA 9.0

It is always desirable to read the official Installation Guide for Linux of the version of CUDA you want to install. In my case, the version is 9.0.

  1. Download a runfile(local) install file from the NVIDA homepage.
sudo sh cuda_9.0.176_384.81_linux.run

While running the file, you are asked to install the NVIDIA driver, CUDA toolkit, and example codes. The example codes are just optional. If you already installed a recent NVIDIA driver, the CUDA toolkit is the only thing you should install. Pick the default choices while running.

Do you accept the previously read EULA?
accept/decline/quit: accept

Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 384.81?
(y)es/(n)o/(q)uit: y

???Something yes or default choice

Install the CUDA 9.0 Toolkit?
(y)es/(n)o/(q)uit: y

Enter Toolkit Location
 [ default is /usr/local/cuda-9.0 ]:

Do you want to install a symbolic link at /usr/local/cuda?
(y)es/(n)o/(q)uit: y

Install the CUDA 9.0 Samples?
(y)es/(n)o/(q)uit: n

  1. Set the enviroment variables PATH and LD_LIBRARY_PATH.

  2. Then,

sudo apt-get install cuda 
reboot