Login

Seminar on Efficient Programming of HPC Systems - Frameworks and Algorithms

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

Hörsaal (Ground Floor)

MPCDF
    • 1
      Introduction
      Speaker: Erwin Laure (MPCDF)
    • 2
      HPX - A modern C++ task parallelization framework
      Speaker: Guillermo Marcos Lara
    • 3
      Portable GPU Programming with OpenMP Target Offloading
      Speaker: Denys Myshak
    • 4
      Julia for High-Performance Computing
      Speaker: Daniel Singh
    • 10:20
      Break
    • 5
      Seminar Paper on HPC Storage and Lustre
      Speaker: Daymon Schodits
    • 6
      Data Layouts on Heterogeneous Systems
      Speaker: Lukas Englhauser
    • 7
      IO-Aware Attention Across GPU Generations: How FlashAttention Tracks the Moving Bottleneck
      Speaker: Tom Osterfeld
    • 11:50
      Lunch break
    • 8
      Alpaka
      Speaker: Bora Uygar Özyurt
    • 9
      An MLIR-Based Approach to HPC Portability
      Speaker: Yannick Schürmann
    • 10
      JAX/XLA for Accelerator Performance Portability: Compiler Mechanisms and Quantitative Evidence
      Speaker: Pau Marín Roig
    • 14:00
      Break
    • 11
      Mitigating Load Imbalance in Molecular Dynamics through Adaptive Task Parallelism
      Speaker: Audrey Kyrene Chen Kartamihardjo
    • 12
      ADIOS in High-Performance Computing Architecture and Performance Improvements for Scientific Simulations
      Speaker: Thomas Krachten
    • 13
      Balancing Performance and Accuracy: Mixed-Precision Algorithms for Linear Algebra in HPC
      Speaker: Felix Weißleder
    • 15:30
      Break
    • 14
      Non-IEEE and Reduced-Precision Number Formats for AI and HPC
      Speaker: Paul Fleischmann
    • 15
      Model parallelization strategies for inference and training
      Speaker: Nils Marvin Quiring
    • 16
      Scaling Massive Models Efficiently: Fully Sharded Data Parallelism in PyTorch
      Speaker: Sai Krishna Sriyash Kommalapati
    • 17
      Conclusions