| Time | Track 1 | Track 2 | Track 3 | ||
| PT AI 1 | PT Arch |
BOS Efficiency | |||
| 9:15 | Machine Learning for Flow Predictions: Rakesh Sarma | JSC + R-CCS | A Multi-Level Post-mortem Analysis of Blue Waters Storage System: François Tessier | Inria | Current Trends in Energy Efficiency for HPC: Solomon Bekele-ANL, Power Monitoring on Aurora; Julita Corbalan (tentative)- BSC EAR; Thomas Breuer-JSC LLview |
| 9:35 | Applications of Rehearsal-based Continual Learning to Generative Tasks, Numerical Simulations and Beyond: Alexandru Costan, Bogdan Nicolae | Inria + ANL | Updates on Deployment of ORCHA in Flash-X: Youngjun Lee, Anshu Dubey | ANL | |
| PT AI 1 | PT Arch | BOS Efficiency | |||
| 10:15 | Do Large Language Models Speak Scientific Workflows? Orcun Yildiz | ANL | Shared Infrastructure for Source Transformation Automatic Differentiation: Jan Hückelheim | ANL | Current Trends in Energy Efficiency for HPC: Guillaume Pallez-INRIA Extending the livetime of HPC systems, Keiji Yamamoto-RIKEN Energy efficiency for Fugaku |
| 10:35 | On the Reproducibility Challenges of Federated Learning: Investigating the Gap between Simulation, Emulation and Real-World Deployments: Kate Keahey | ANL | Porting StarPU on top of nOS-V the hypervisor for enabling runtime system interoperability: Olivier Aumage | Inria | |
| PT Res | PT Apps | ||||
| 11:15 | Fault-tolerant numerical iterative algorithms at scale: Yves Robert | Inria | Optimizing SERGHEI-SWE flood simulations with malleable resources matching flood dynamics: Daniel Cavieder-Voulliéme | JSC | |
| 11:35 | Towards Affordable Reproducibility Using Scalable Capture and Comparison of Intermediate Multi-Run Results: Kevin Assogba | ANL | Leveraging Idle CPU Resources to Mitigate GPU Contention on the Example of NekRS: Victor Mateevitsi | ANL | |
| K3 | |||||
| 12:00 | TBA | ||||
Choose timezone
Your profile timezone:
