Program

CANOPIE-HPC 2025 Workshop Program

Monday, November 17th, 2025 - 9:00AM-5:30PM CST

Room 275

Note all times listed are in CENTRAL STANDARD TIME (UTC–5)

Agenda

Start	Duration	Title	Speaker	Type
09:00 AM		Start
09:00 AM	15 min	CANOPIE-HPC - Introduction and welcome	Madonna, Misale, Canon, Younge
09:15 AM	45 min	The Convergence of Containers and Kubernetes with HPC and Generative AI	Gupta, Jackson. Red Hat	Invited Talk
10:00 AM	30 min	Morning Break
10:30 AM	30 min	Experience Deploying Containerized GenAI Services at an HPC Center	Beltre, Ogden, Pedretti. Sandia National Laboratories	Paper
11:00 AM	30 min	Engine-Agnostic Model Hot-Swapping for Cost-Effective LLM Inference	Stoyanov (University of Oxford, Red Hat), Spišaková, Reber, Armour, Copik, Bruno	Paper
11:30 AM	10 min	To Virtualize or Not to Virtualize: Experiences from Building Two Generations of Virtualized Infrastructure for LLM Training	Mohan, Chen, Chung, Welp, Seelam. IBM Research	Lightning talk
11:40 AM	10 min	Dynamic Topology-Aware Scheduling in HPC Systems with Topograph	Shmulevich. NVIDIA Corporation	Lightning talk
11:50 AM	10 min	Evolving Sarus to augment Podman for HPC	Madonna, Cruz, Lee, Chesi. ETH Zurich, CSCS	Lightning talk
12:00 PM	30 min	Seamless End-to-End Containerized HPC Environments	Cook, Canon, Lavely, Margala. NERSC, Lawrence Berkeley National Laboratory	Paper
12:30 PM	90 min	Lunch Break
2:00 PM	30 min	Usability Evaluation of Cloud for HPC Applications	Sochat, Milroy, Sarkar, Marathe, Patki. Lawrence Livermore National Laboratory	Paper
2:30 PM	30 min	An Elastic Job Scheduler for HPC Applications on the Cloud	Bhosale, Chandrasekar, Kale, Kokkila-Schumacher. University of Illinois at Urbana-Champaign, IBM Research	Paper
3:00	30 min	Afternoon Break
3:30 PM	10 min	Charliecloud with Nvidia GPUS via standards new and old	Phinney, Loshak. Los Alamos National Laboratory	Lightning talk
3:40 PM	10 min	Charliecloud as a Kubernetes Container Runtime	Loshak, Bielicke. Los Alamos National Laboratory	Lightning talk
3:50 PM	30 min	Towards Enabling Hostile Multi-tenancy in Kubernetes	Kanso, Oks, Mostafa, Virdi. Microsoft Corporation	Paper
4:20 PM	30 min	Evaluating HPK for Running Cloud-Native Workloads on Slurm Clusters	Chazapis, Vassilakis, Petsis, Marazakis, Bilas. Foundation for Research and Technology - Hellas	Paper
4:50 PM	40 min	Community Discussion and Open Q&A	Madonna, Misale, Canon, Younge	Panel
5:30 PM		End