Introduction to Node-Level Parallel Programming

Performance Engineering Course

Europe/Berlin
Virtual (Zoom)

Virtual

Zoom

Description

Topics covered in this two days course include:

  • Access the HPC cluster systems at PC²
  • Basics of CPU microarchitecture
  • Introduction to OpenMP
  • SIMD vectorization
  • Microbenchmarking on Noctua
  • Parallel programming on ccNUMA architecture
  • Roofline Model
  • Hardware performance counter with LIKWID
  • Performance engineering with patterns
  • Case study: dense matrix multiplication
  • Case study: miniMD

This course also includes hands-on sessions on the HPC cluster system at PC².

Please note:

  • The course assumes:
    • basic knowledge of the Linux shell. We provide a short online selftest. In case you have no experience with Linux, we recommend that you attend our next Linux Introduction Course.
    • basic knowledge of workload managers (esp. Slurm) and jobscripts. We provide a short online selftest. In case you have no experience with workload manager on clusters, we recommend that you attend our next HPC Introduction Course.
    • knowledge of HPC system architecture, profiling and scaling analysis. Additionally, a little bit of programming knowledge (C/C++ or Fortran) is required. We provide a short online selftest. In case you have no experience with the architecture of modern HPC compute nodes, we recommend that you attend our next course on Advanced Topics of HPC.
  • The language of the course is planned to be English. If you don't feel comfortable enough with English, we can make this course also in German at a later date upon request.
  • For the hands-on session of the course no special operating system is required. You can use your own computer with Windows, macOS or Linux.
  • This course will be organized online by using the teaching system at Paderborn University due to the concern of the Corona virus. Once your registration has been confirmed, more detailed information will be available.

This course is free of cost for members of German universities or publicly-funded research institutions in Germany.

Organized by

PC² HPC advisors