Getting Started with Containers on HPC through Singularity
AI/Machine Learning/Deep Learning
Education and Training
Scientific Software Development
System Software & Runtime Systems
TimeSunday, June 24th9am - 1pm
DescriptionWithin just the past few years, the use of containers has revolutionized the way in which industries and enterprises have developed and deployed computational software and distributed systems. The containerization model is gaining traction within the HPC community as well with the promise of improved reliability, reproducibility, and levels customization that are not otherwise possible on supercomputers previously. From the onset of containerization in HPC, Singularity has lead the way in providing container services, ranging from small clusters to entire supercomputers.
This tutorial looks to educate users on the usability of containers on HPC resources through the use of Singularity. Specifically, the tutorial will walk through the processes of creating, deploying, and running Singularity containers on a linux cluster. We will provide a detailed background of containers, along with introductory hands-on experience building and running containers on an cluster. Furthermore, the tutorial will provide more advanced information on how to run MPI-based HPC applications as well as cutting-edge machine learning toolkits. Users will leave the tutorial with a solid foundational understanding of how to utilize containers with HPC resources through Singularity, as well as in-depth knowledge to deploy custom containers on their own resources.
Content Level 40% beginner, 30% intermediate, 30% advanced
Target Audience This meeting aim is the education and formation of computer science researchers, IT professionals, system administrators, computer science researchers and engineer students, able to propose solutions for distributed systems and HPC problems in their projects, or whose research will use/develop HPC or cloud computing applications.