Mr Aidan Wilson1
1Intersect Australia, Sydney, Australia
Biography:
Aidan Wilson is Digital Research Services Manager at Intersect Australia, where he supervises a team of Digital Research Analysts, and manages Intersect’s researcher training program. Aidan has been heavily involved in training delivery and systems since he joined Intersect in 2015.
Prior to joining Intersect, Aidan was a linguistic researcher working in the field of Australia’s Indigenous Languages, focusing in particular on the verb morphology of Traditional Tiwi, and other languages from the top-end, one of the most linguistically diverse regions on Earth.
http://orcid.org/0000-0001-9858-5470
Abstract:
Access to software can be a barrier for many people in attending and fully participating in digital skills training. Unless the learner spends time preparing for the course by installing software and downloading data, trainers spend time during the course ensuring that everyone is ready to participate.
For years, we have been developing and refining systems that enable frictionless training where attendees (in programming courses in particular) have access to all necessary tooling: software and data. The latest iteration of these efforts, Trainstation, has eliminated account creation and course setup issues for all programming courses, and has been central to our training delivery for nearly two years.
This year, we turn to address the problems presented by HPC training delivery: In many cases, HPC administrators expect users to have taken training in HPC before they can be granted access to top-tier systems. But how can users take training without access to the cluster? To break this reciprocal dependency, we have developed a virtualised Slurm cluster using Docker, for delivery of HPC courses in a sandboxed environment. Integration with Trainstation furthermore ensures that user accounts, data, and HPC resources are provisioned and ready with minimal manual effort.
This presentation will focus on these two case studies—Programming courses, such as Python, R; and High-Performance Computing courses—the different challenges these courses present for training delivery, how a system like Trainstation can address these challenges, and how we might deploy similar solutions for other courses.