Cluster System Administration for Non-Sysadmins (2:00-6:00)
Diversity Equity Inclusion (DEI)
Education and Training and Outreach
HPC Training and Education
Parallel Programming Languages and Models
TimeMonday, 15 November 20212pm - 6pm CST
DescriptionHave you ever wondered what goes into administering an HPC system? This is your chance to learn about it! This workshop will cover the basic concepts and services that are involved in running a high-performance computing cluster. Topics will include ethernet and networking protocols, hostname management, automatic booting of compute nodes, unintended stateful node installs, stateless node images, firewalls, user management, shared file systems and resource managers.
The presenters will provide a small cluster utilizing the Extreme Cluster Administration Toolkit (xCAT) that attendees can log into and explore. The workshop will be primarily based on demonstrations with information provided to attendees via primary sources of documentation rather than slides. By the end of the workshop, attendees will understand the basic terms and functions of the various services that make up a compute cluster and where to go to find more information. The presenters will also provide a GitHub that has a starting xCAT configuration and other service configuration files for a simple cluster. These configuration files can be used in the future by attendees as a starting point for building their own clusters at their home institutions.