Singularity

Singularity is a popular container solution for HPC systems. It natively supports GPU and high performance networks in containers and is compatible with docker images. More info at: https://sylabs.io/docs/

Grid'5000 supports the Singularity containers: Singularity is available in the standard environment and it does not requires to run it as root.

Basic usage

Just run the singularity command to use it. It can also be run in a OAR submission (none-interactive batch job). For instance:

frontend:

oarsub -l core=1 "/grid5000/code/bin/singularity run library://sylabsed/examples/lolcow"

	Note
	The full path to /grid5000/code/bin/singularity is required for non-interactive OAR jobs

The Singularity user documentation is available at https://sylabs.io/guides/3.5/user-guide. It describes the various ways to run programs inside a container and how to build your own container image (which requires to be root, but can be performed on your own laptop or on a Grid'5000 node using "sudo-g5k").

Using docker containers with Singularity

Singularity can also be used to start docker containers. For instance:

node:

singularity run docker://gentoo/stage3-amd64

Running MPI programs in Singularity containers

MPI programs may be run in Singularity containers, by leveraging both the MPI implementation available in the host, i.e. a Grid'5000 physical node (which has a direct access to the high peformance network hardware if present), and the MPI library that must be installed inside the container.

MPI programs in the Singularity container can then be started using the the mpirun command on the host.

See https://sylabs.io/guides/3.5/user-guide/mpi.html for more information.

For instance, to submit such a MPI job under OAR, assuming your avec a Singularity image named my_mpi_image.sif in your home directory, use:

frontend:

oarsub -l nodes=2 "mpirun -hostfile \$OAR_NODE_FILE --mca orte_rsh_agent oarsh -- /grid5000/code/bin/singularity exec my_mpi_image.sif /opt/mpitest"

Using GPUs in Singularity containers

GPUs available in the host can be made available inside the container by using the --nv option (for Nvidia GPUs only).

For instance, to start an interactive tensorflow environment with one GPU, first submit the job reserving 1 GPU:

frontend:

oarsub -I -l gpu=1

Then on that node:

node:

singularity run --nv docker://tensorflow/tensorflow:latest-gpu

More info at: https://sylabs.io/guides/3.5/user-guide/gpu.html