This full day in-person event will include presentations, demos, hands-on exercises, networking opportunities and a tour of IU’s datacenter. Lunch and breaks will be provided, all at no cost to attendees. A recording of the presentations and downloads of the hands-on exercises will be available for those unable to attend in-person.
Objectives
Attendees will:- Learn about Big Red 200 and the latest tools from NVIDIA to take advantage of the GPU enabled nodes on Big Red 200.
- Receive hands-on Deep Learning exercises to help get acquainted with using Big Red 200.
- Network with Research Technologies and NVIDIA personnel as well as with other IU researchers working on Big Red 200.
- Learn about other services and systems supported and maintained by Research Technologies and the people who support them.
Prerequisites
- An account on BigRed200 (requested 5 days prior to day of the event). (https://kb.iu.edu/d/brcc#access)
- Basic knowledge of Unix and Python.
- A means to login via a terminal.
- Recommended (but not required): Familiarity with a text editor.
Go to workshop materials on GitHub
Oct 20 Big Red 200 and AI Day materialsRecordings available on YouTube
Welcome by Matt LinkAgenda
- 10:00: Welcome by Matt Link, Associate Vice President for UITS Research Technologies (RT) (10 minutes)
- 10:10: HPC stories from the trenches (60 minutes)
- Enrico Vesperini/Vaclav Pavlik, Department of Astronomy - IUB (30 minutes)
- Michael Robo, IU School of Medicine (30 minutes)
- 11:10: Intro to HPC at IU with Laura Huber, Research Software Developer, RT (50 minutes)
- Brief overview
- Accessing system
- Accessing packages and software
- Noon: Lunch and networking (60 minutes)
- 1:00: Intro to DL with Scott Teige, Research Software Developer, RT (75 minutes)
- 2:15: 2:00: Break (15 minutes)
- 2:30: 2:15: NVIDIA presentations with Kaleb Smith, Sr. Data Scientist and Marc West, Sr. Account Manager (90 minutes)
- NGC containers
- Nemo/NLP
- Vision AI
- RAPIDS
- 4:00: 3:45: Closing remarks by Abhinav Thota, Manager, Research Applications and Deep Learning and Scott Michael, Director, Research Software and Solution, RT
- 4:05: 4:00 Optional data center tour with Matt Allen, Manager, High Performance Systems, RT
The Supercomputing for Everyone Series (SC4ES) aims to bring more users into the realm of advanced computing, whether it be visualization, computation, analytics, storage, or any related discipline. Research Technologies can take you to the next level of computing.
Supercomputing for Everyone Series workshops and seminars are led by personnel from Research Technologies, a division of University Information Technology Services and a center in the Pervasive Technology Institute at Indiana University.