AI Training Series: SmartSim at OLCF
SmartSim at OLCF
Thursday, July 13, 2023
While HPC support for AI has grown quickly over the past few years, AI support for HPC has not. Integrating Machine Learning into HPC simulations represents a challenging engineering task. SmartSim was created to ease this burden on scientists.
SmartSim is a workflow library that makes it easier to use common Machine Learning (ML) libraries, like PyTorch and TensorFlow, in High Performance Computing (HPC) simulations and applications. This allows the launching of ML infrastructure on HPC systems alongside user workloads. SmartSim also allows ML to be written in Python, while still interfacing with Fortran, C, and C++ applications at HPC scale.
In this training, an overview of SmartSim and best practices will be provided, along with specific examples of how to use it on OLCF systems. This training is intended for current OLCF users attached to Moderate Security Enclave projects (able to access systems like Summit, Frontier). Non-OLCF users are welcome to sign-up, attend the virtual presentations, and encouraged to download the public tutorials for themselves (see GitHub link below).
|1:00 pm – 1:10 pm EDT||Welcome and Introductions from SmartSim Team||Matt Ellis, Andrew Shao, Matt Drozt (HPE)|
|1:10 pm – 2:00 pm EDT||Overview of SmartSim and Use Cases||Matt Ellis, Andrew Shao, Matt Drozt (HPE)|
|2:00 pm – 2:45 pm EDT||Hands-on Frontier w/ MOM6 Ocean Model||Matt Ellis, Andrew Shao, Matt Drozt (HPE)|
|2:45 pm – 3:00 pm EDT||Open Q&A / Discussion||Matt Ellis, Andrew Shao, Matt Drozt (HPE)|
**Registration closed (as of 5PM Eastern July 12th)**
Joining information will be sent to you in a calendar event before the event.