OLCF Hands-On Workshop
Thursday, September 3, 2026 · 2:00–5:00 p.m. ET
Presented by
Andrew Shao, HPE;
Matt Drozt, HPE; and
Aymen Alsaadi, Rutgers University
This event is co-hosted by OLCF, ALCF, and NERSC.
OVERVIEW
Many of the underlying techniques that drive today’s AI applications can be repurposed for advancing scientific simulations. This workshop focuses on two classes of hybrid AI/simulation workflows: AI-in-the-loop where the AI component is coupled tightly within the simulation’s inner loop (e.g. for sub grid-scale modelling) and AI-around-the-loop where AI techniques are used to drive a simulation (e.g. for parameter exploration). Scientists who are looking to include these as a tool for scientific research run into a variety of challenges including how to deal with the heterogeneous nature of the compute, passing data and information between components, and how to define the workflows themselves which cannot be composed using existing software. To address these difficulties, this workshop introduces RHAPSODY, a tool co-developed by Rutgers University and HPE which is being used across the Genesis Mission to accelerate scientific discovery. The first part of the workshop will be a lecture that will outline RHAPSODY, discuss HPE’s DragonHPC backend, and also show a variety of scientific uses cases that were conducted using HPE’s SmartSim (one of the predecessors to RHAPSODY). The second part of the workshop will be hands-on and show how RHAPSODY can be used to drive hybrid applications using OpenFOAM and other simulation codes. Additional time will be allotted for any researchers who are interesting in adapting these workflows to their use cases.
Topics Covered
Hybrid AI and simulation workflows
AI-in-the-loop
AI-around-the-loop
RHAPSODY
DragonHPC
SmartSim use cases
Heterogeneous computing
OpenFOAM workflows
Workshop Format
This workshop includes both a lecture and a hands-on technical session. Participants will also have an opportunity to discuss adapting RHAPSODY workflows to their own research applications.
Hands-On Access Options
Existing users of participating HPC facilities may use their current allocations to follow along with the hands-on exercises. Participants without an existing facility account may follow along on their personal computers or request access to a training account through a dedicated training project on
Odo.
Additional instructions for accessing the Odo training project will be provided to registered participants.
