Overview
On cluster newton installed environment enroot to use a containers (dockers and etc. )
Before you can use containers we have to create it (if not exist ) , so if you can’t find needed container in /mnt/images you can send request mail to admlab@cs.technion.ac.il with : full name and path to docker/singularity container
To use containers :
Load module enroot:
module purge module load enroot
Copy needed image to home directory:
mkdir /enroot cp /mnt/images/nvidia+cuda+10.0-base.sqsh enroot/ cd /enroot
Open image to enroot local environment:
enroot create --name cuda nvidia+cuda+10.0-base.sqsh
* for check loaded images run command:
enroot list
Request resources to run container:
srun -c 10 --gres=gpu:2 enroot start cuda
Examples
Starting a container
... enroot start --rw nvidia+pytorch+21.04-py3 bash
Start the container nvidia+pytorch+21.04-py3 in read-write mode (–rw) and run bash inside the container.
... enroot start --root --rw nvidia+pytorch+21.04-py3 bash
Start container in –rw-mode and get root access (–root) inside the container.
You can now install software with root privileges, depending on the containerized Linux distribution e.g. with apt-get install …, apk add …, yum install …, pacman -S …
... enroot start -m :/work --rw nvidia+pytorch+21.04-py3 bash
Start container and mount (-m) a local directory to /work inside the container.
... enroot start -m :/work --rw nvidia+pytorch+21.04-py3 jupyter lab
Start container, mount a directory and start the application jupyter lab.