Armando Acosta, Dell Technologies, and Alan Chalker, Ohio Supercomputer Center | SC23
summary
State-funded Ohio Supercomputer Center aggregates four Dell clusters—about 60,000 cores and roughly $25,000,000 in hardware—and runs Open OnDemand which now appears at nearly 700 sites in 62 countries. Art students launch Stable Diffusion on shared NVIDIA V100 GPUs inside Kubernetes pods, showing generative AI + HPC for training and low-latency inference. Discussion highlights concrete needs: model governance, reproducibility, data provenance, and using validated designs and standards to lower barriers and scale real workloads.
AI-written summary · about
chapters (17)
- 0:00Welcome to SC23
- 0:43AI and HPC power dynamic
- 1:03Ohio Supercomputer origin
- 2:06Statewide compute benefits
- 3:08Dell HPC community event
- 3:50HPC and AI integration
- 4:33Open OnDemand demo
- 7:21Open OnDemand history
- 9:25Open OnDemand adoption
- 10:29New enablement use cases
- 11:53AI help app in OnDemand
- 13:17AI as a service future
- 14:20Democratization of compute
- 15:45Ecosystem changes
- 18:51AI iterative workflows
- 20:34Future of Open OnDemand
- 21:16Closing remarks
