Login

Seminar on Efficient Programming of HPC Systems - Frameworks and Algorithms

→ Europe/Berlin
Hörsaal (Ground Floor) (MPCDF)

Hörsaal (Ground Floor)

MPCDF
    • 09:00 → 09:05
      Introduction 5m
      Speaker: Erwin Laure (MPCDF)
    • 09:05 → 09:30
      HPX - A modern C++ task parallelization framework 25m
      Speaker: Guillermo Marcos Lara
    • 09:30 → 09:55
      Portable GPU Programming with OpenMP Target Offloading 25m
      Speaker: Denys Myshak
    • 09:55 → 10:20
      Julia for High-Performance Computing 25m
      Speaker: Daniel Singh
    • 10:20 → 10:35
      Break 15m
    • 10:35 → 11:00
      Seminar Paper on HPC Storage and Lustre 25m
      Speaker: Daymon Schodits
    • 11:00 → 11:25
      Data Layouts on Heterogeneous Systems 25m
      Speaker: Lukas Englhauser
    • 11:25 → 11:50
      IO-Aware Attention Across GPU Generations: How FlashAttention Tracks the Moving Bottleneck 25m
      Speaker: Tom Osterfeld
    • 11:50 → 12:45
      Lunch break 55m
    • 12:45 → 13:10
      Alpaka 25m
      Speaker: Bora Uygar Özyurt
    • 13:10 → 13:35
      An MLIR-Based Approach to HPC Portability 25m
      Speaker: Yannick Schürmann
    • 13:35 → 14:00
      JAX/XLA for Accelerator Performance Portability: Compiler Mechanisms and Quantitative Evidence 25m
      Speaker: Pau Marín Roig
    • 14:00 → 14:15
      Break 15m
    • 14:15 → 14:40
      Mitigating Load Imbalance in Molecular Dynamics through Adaptive Task Parallelism 25m
      Speaker: Audrey Kyrene Chen Kartamihardjo
    • 14:40 → 15:05
      ADIOS in High-Performance Computing Architecture and Performance Improvements for Scientific Simulations 25m
      Speaker: Thomas Krachten
    • 15:05 → 15:30
      Balancing Performance and Accuracy: Mixed-Precision Algorithms for Linear Algebra in HPC 25m
      Speaker: Felix Weißleder
    • 15:30 → 15:45
      Break 15m
    • 15:45 → 16:10
      Non-IEEE and Reduced-Precision Number Formats for AI and HPC 25m
      Speaker: Paul Fleischmann
    • 16:10 → 16:35
      Model parallelization strategies for inference and training 25m
      Speaker: Nils Marvin Quiring
    • 16:35 → 17:00
      Scaling Massive Models Efficiently: Fully Sharded Data Parallelism in PyTorch 25m
      Speaker: Sai Krishna Sriyash Kommalapati
    • 17:00 → 17:20
      Conclusions 20m