Scientific applications are increasingly adopting Artificial Intelligence (AI) techniques to advance science. There are specialized hardware accelerators designed and built to run AI applications efficiently. With a wide diversity in the hardware architectures and software stacks of these systems, it is challenging to understand the differences between these accelerators, their capabilities, programming approaches, and how they perform, particularly for scientific applications.
In this tutorial, we will cover an overview of the AI accelerators landscape focusing on SambaNova, Cerebras, Graphcore, Groq, and Habana systems along with architectural features and details of their software stacks. We will have hands-on exercises to help attendees understand how to program these systems by learning how to refactor codes and compile and run the models on these systems. The tutorial will provide the attendees with an understanding of the key capabilities of emerging AI accelerators and their performance implications for scientific applications
| Date | 17 November 2024 |
|---|---|
| Time | 8:30am - 5pm EST |
| Location | B201 |
| Time (EST) | Topic/ Speaker |
|---|---|
| 08.30 - 08.40 AM | Programming Novel AI Accelerators for Scientific Computing Murali Emani (ANL) [Slides] |
| 08.40 - 09.10 AM | The Power of Wafer-Scale Systems Cerebras HPC Research and SDK Overview Leighton Wilson (Cerebras) [Slides AI] [Slides SDK] |
| 09.10 - 10.00 AM | Hands-on With Cerebras Systems Leighton Wilson (Cerebras), Varuni Sastry (ANL) [Instructions] |
| 10.00 - 10.30 AM | Coffee Break |
| 10.30 - 11.00 AM | Programming Intel Habana Gaudi 2 for Scientific Computing Buke Ao (Intel Habana) [Slides] |
| 11.00 - 11.30 AM | Hands-on with Intel Habana Gaudi2 Buke Ao (Intel Habana) [Instructions] |
| 11.30 - 12.00 PM | GenAI Training and Inference at Scale with SambaNova Systems Petro Junior Milan (Sambanova) [Slides] |
| 12.00 - 01.30 PM | Lunch Break |
| 01.30 - 02.00 PM | Hands-on with Sambanova Systems Petro Junior Milan (Sambanova), Varuni Sastry (ANL) [Instructions] |
| 02.00 - 02.30 PM | Groq’s approach to HW/SW Systems for LLM Inference Sanjif Shanmugavelu (Groq) [Slides] |
| 02.30 - 03.00 PM | Hands-on with Groq Systems Sanjif Shanmugavelu (Groq), Varuni Sastry (ANL) [Instructions] |
| 03.00 - 03.30 PM | Coffee Break |
| 03.30 - 04.00 PM | Introduction to Graphcore's IPU and Programming model : Sid Raskar (ANL) [Slides] |
| 04.00 - 04.30 PM | Hands-on with Grapchore Systems Sid Raskar (ANL) [Instructions] |
-
Request an ALCF Computer User Account if you do not currently have one
-
If you have an ALCF Account that is currently inactive, submit an account reactivation request*.
-
If you have an active ALCF account, click Join Project to submit a membership request.
Contact accounts@alcf.anl.gov M-F 9am to 5pm CT.
Reach out to us on slack channel #help-accounts on ALCF-AIAccelerator-tutorials Slack.
SC24 Tutorial allocation will stay active till end of November 2024.
To gain access to AI Testbeds at ALCF after tutorial allocation expires apply for Director’s Discretionary Allocation Program
The ALCF Director’s Discretionary program provides “start up” awards to researchers working to achieve computational readiness for for a major allocation award.
- SC24 Tutorial Webpage
- Github Tutorial Repository
- Overview of AI Testbeds at ALCF
- ALCF AI Testbed Documentation
- Director’s Discretionary Allocation Program
- Join Slack Workspace
Contributors: Siddhisanket (Sid) Raskar, Varuni Sastry, Bill Arnold, Murali Emani.
This research used resources of the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC02-06CH11357.