The rational
The computational performance of super computers is usually expressed in terms of their theoretical peak performance or measured with rather simple and easy to handle benchmark codes, like the High-Performance Linpack Benchmark, a measure of the floating-point computing power. These metrics allow to easily rank the performance of computers but have only limited value, when it comes to the question how well a computer system is suited for specific complex applications like weather and climate models.

The EU funded projects ESCAPE and ESiWACE therefore developed the High Performance Weather and Climate Benchmark, short HPCW, which consists of widely used European weather and climate models and standalone model components which are less complex than the models itself but represent selected characteristics of the models. Our ambition is  to establish HPCW as a useful tool for the communities of weather and climate modellers on one hand and computer manufactures on the other hand.
Currently HPCW consist of:
  1. The HPCW framework
  2. Numerical models
    • ICON - Ocean and Atmosphere
    • IFS - Atmosphere (RAPS version)
    • NEMO - Ocean
  3. Mini-applications (dwarfs):
    • IFS atmosphere FV dwarf IFS-FVM
    • Radiation dwarf ACRANEB2
    • ICON ocean advection dwarf
    • ECRad
    • ECTrans
    • Dwarf-P-CloudSC
  4. Features
    • Verification procedures for automatic correctness checking
    • Test-cases per model (small, medium, big) for scalability studies
    • Automatic performance metrics extraction (like time to solution)

References for further reading:

A talk by David Guibert (Eviden, 2023): HPCW - A Framework for Reproducible Benchmarks

The ESIWACE2 deliverable describing HPCW v1. (Biercamp, et al, 2022)