Container tools for HPC, Singularity, Docker.
The introduction of container technologies has deeply modified the way in which software stacks
can be developed, deployed, tested, shared and executed in production environments. Although such a revolutionary way of work was initially used in enterprise and company, it is being now adopted in High Performance Computing environments also. Many containerization tools suitable for High Performance Computing systems have been developed in the last years, such as Singularity, Shifter, Charliecloud, Podman, Sarus and so on. All of them allow researchers to deploy a complete software stack for their workflows, achieving flexibility, reliability, ease of deployment, and portability that containers allow.
Through this course, participants will be introduced into the possibility of building and running both their own or pre-built applications and software stacks using HPC containerization tools.
The course will be mainly focused on Singularity. Some information covering other containerization tools will be also provided.
The course spans in 4 morning days. Theoretical lessons will be interspersed with exercise sessions to fix the concepts explained.
PRACE users, final year master students, PhD students, and young researchers in
computational sciences and engineering, with different backgrounds, interested in applying the
containerization technologies on high performance computing to their research.
Unix-Linux operating system knowledge; software installation on linux-unix OS; experiences in running applications on HPC clusters