RIT Scientific Computing Group

I run an informal seminar/workshop series on campus on the subject of scientific computing, for anybody interested. We meet every Monday at 1pm in Orange Hall 1350. Check out the website and poster for more info, including upcoming talks and an archive of past talks.


1 Like

Just curious, do y’all cover many topics around management of scientific computing infrastructure? Or is it mostly focused around the programming aspects? I worked with some Linux supercomputers the past few summers, even if it wasn’t too deep. But I didn’t run many jobs in the supercomputer, just ways to test it or configure it different ways.

We have covered that a little bit in the past, and the reason we haven’t covered more is not for lack of desire. This was probably our most computing infrastructurey talk

If you wanted to give a little overview on how to work with Linux supercomputers, or even just talk about your personal experience, that’d be awesome. Assuming there are others in the room who also have supercomputer experience (depends on who shows up that day, but there are definitely people who have in the group) it could even be more of a dialog.

1 Like

This is some of what my summer looked like:

I could probably give a 15-20 minute talk summary of the different container tools for HPC supercomputers out there. Depending on how much formality is needed (e.g. audience size, amount of prep time for good slides, etc.), I can be ready to go in 2-3 weeks or later next month.

This would actually be awesome! I know a lot of people (at least a few in the group) use Singularity containers. It would be very interesting to see performance comparisons for them. Plenty of others probably haven’t even heard of containers, so they’d probably benefit from introductory slides explaining what they are. The talk can be completely informal, so don’t worry too much about making perfect slides, a much bigger emphasis should be placed on accessibility to a broad audience.

Whenever you have a better sense of what you’d like to present, and when, you can fill out the form at the link below, and I’ll get an email I won’t lose track of with all the information :slight_smile: and we can figure out exactly where to fit it into the schedule from there.


1 Like

To follow up here, come by on Monday, 18 November at 1:00pm in Orange Hall 1350 for Docker containers + supercomputers = ??

Here is the abstract:

Docker containers and related products have revolutionized DevOps, but where do containers belong in supercomputing / HPC infrastructure, on large distributed computing grids? In a world of proprietary hardware and drivers, large-scale distributed systems, and emphasis on bare-metal performance, are containers another virtualization fad to skip over supercomputing? Guess again.

This session explores different container run-times built for supercomputing / HPC environments, what they offer, and some benefits/costs of implementing them alongside HPC job scheduling software. This session will be useful for you if you work with HPC infrastructure, you write code intended to be run on parallel systems, or you are interested for what the future of HPC technology might look like.

Hope to see you there!