Event · D3A 4.0 2026
Supercomputing: Agents & Challenges
Thursday 8 October 2026 · 90 minutes · D3A 4.0 2026, room to be announced.
Shared HPC only pays off when researchers can get their work done on it. This session puts HPC providers and researchers in the same room to name the frictions that are in the way, and to investigate where agentic GenAI opens new possibilities.
About this session
Format
- 90 minutes, interactive
- Talks 33% or less
- Discussion 61% or more
Level
- Intermediate
- HPC experience helpful
- No agent expertise required
Audience
- AI/ML researchers
- HPC operators
- Domain scientists on shared compute
Programme
Three short talks set the stage. The rest of the session is breakouts, plenum, and synthesis.
Framing and synthesis Talk Interactive
| Time | Session | Speaker |
|---|---|---|
| 00:00-00:05 | Welcome and framing | Benjamin Starostka Jakobsen |
| 00:05-00:15 | Agentic HPC: GenAI and sustainability | Claudio Pica SDU eScience Center |
| 00:15-00:25 | Operating HPC: Hosting research at scale | Ulrik N. de Lichtenberg DCAI / Gefion |
| 00:25-00:35 | Shared HPC: Effective sharing of GPUs | Pınar Tözün IT University of Copenhagen |
| 00:35-00:55 | Breakout discussions: HPC providers and researchers | All participants |
| 00:55-01:15 | Plenum: report-back from breakouts | Benjamin Starostka Jakobsen |
| 01:15-01:30 | Synthesis: common findings and next steps | Nicki Skafte Detlefsen |
Outcomes
Questions we hope to answer together.
- Which frictions are blocking researchers from getting work done on shared HPC?
- What should HPC providers pick up next, and what useful practices can researchers share with each other in the meantime?
- As agents take over cluster work, will researchers still need cluster knowledge, and who debugs it when the agent gets it wrong?
Organizers
- Nicki Skafte Detlefsen DTU nsde@dtu.dk
- Ulrik N. de Lichtenberg DCAI ulrik.de.lichtenberg@dcai.dk
- Benjamin Starostka Jakobsen P1 bstja@dtu.dk
Collaborations
Pioneer Centre for Artificial Intelligence Pan-Danish research centre advancing AI, machine learning, and computer vision.
Technical University of Denmark DTU Computing Center provides centralised HPC services, software stacks, storage, and training to researchers across the university.
Danish Centre for AI Innovation Home to Gefion, Denmark's NVIDIA DGX SuperPOD (191 DGX H100 nodes, 1,528 H100 GPUs) for large-scale AI projects.