Other Results
Other Results
Other Results: Within ESiWACE3, Other Results include a portfolio of software tools, model codes, and applications developed to advance performance assessment and optimisation for Earth system modelling. These results support researchers and practitioners in analysing computational efficiency, testing scalability, and improving portability across emerging HPC architectures.
This section highlights these Other Results and the results they generate, showcasing practical solutions built around realistic climate and weather workloads. By focusing on representative scientific use cases rather than synthetic metrics such as High Performance Linpack (HPL), they provide actionable insight into how computing systems perform for real modelling and forecasting applications.
- Ported version to RISC-V Linux of the HPCW suite.
Website Update
Updating the website regularly with information, news, events and materialOphidia accessing Zarr format
Zarr format support for the Ophidia HPDA framework and related assessment. Based on the NCZarr plugin, this extension allows to handle Zarr datasets, reducing data access time while introducing low overhead.Summer School 2024
Training of Early Career Scientists in methods, tools and theoretical knowledge needed to effectively use the latest HPC environments in their research, and to support students in building their network. Additionally, the teaching material will be published for public use.Summer School 2025
Training of Early Career Scientists in methods, tools and theoretical knowledge needed to effectively use the latest HPC environments in their research, and to support students in building their network. Additionally, the teaching material will be published for public use. BoldICON-HPCW
This is the application of an existing code (the Climate and Weather Model ICON. NOT developed by ESiWACE3) within the HPCW Benchmark suite.Training Calendar
Collecting and publishing domain-specific activities such as ESM user forums, workshops, and summer schools on topics of computational and data sciences as well as high performance computingEfficent data layouts for ESM output
data layouts for fast ingestion, cataloguing, distribution and retrievalESiWACE3 Service 1
Service 1 creates small collaborative projects between ESiWACE3 partners and researcher in weather and climate to provide guidance, assistence, and training to help researchers with porting and optimizing their weather and climate simulations for modern HPC systems.Energy-Efficient GPU Computing SC23 tutorial
Training researchers and practitioners how to use improve the energy efficiency of their GPU applications. Demonstrating how to use Kernel Tuner and Kernel Float to this end.Porting Integration EC-Earth4
Porting/integration on LUMI and national systems (MN5). Working in collaboration with EC-Earth ConsortiumService catalogue
Collection of partner expertise and specific service offers for the publicCartopy-Pyodide
Pyodide is a Python runtime in WebAssembly that allows Python to run within your webbrowser in reproducible environments without requiring any installation. Cartopy is a key Python library for weather and climate research to visualise geographical data. Our contribution is to port Cartopy to work inside Pyodide and to contribute it to the Pyodide project such that Cartopy can now be used in Pyodide environments by anyone.Compression-Hackathon
The compression of model states while running a simple toy model (Lorenz96) was explored. The model was implemented on LUMI GPUs and tested with ZFP compression on the GPU. The effect of not compressing every model state but regular restart states was investigated. Results of the ESiW3 hackathon in October 2023.Field Compression Benchmark Suite
The field compression benchmark provides a highly-scalable environment to benchmark and analyse the performance of different data compression methods on a large and diverse set of weather and climate datasets. The Field Compression Benchmark Suite laid the technical groundwork for the ClimateBenchPress benchmark project, by Tim Reichelt (Embed2Scale) with Juniper Tyree (University of Helsinki-ESiW3), PetersXXXX (XXXX-ESiW3) and SaraXXXX (XXXX-ESiW3) as co-authors.NetCDF-Pyodide
Pyodide is a Python runtime in WebAssembly that allows Python to run within your webbrowser in reproducible environments without requiring any installation. NetCDF4 is a key Python library for weather and climate research to work with the NetCDF data format. Our contribution is to port NetCDF to work inside Pyodide and to contribute it to the Pyodide project such that NetCDF can now be used in Pyodide environments by anyone.