New Methodologies for Computational Performance Evaluation of Climate and Weather Models
| Main Authors: | Mario Acosta, Oriol Tintó, Kim Serradell, Stella Paronuzzi, Miguel Castrillo |
|---|---|
| Format: | info Proceeding |
| Bahasa: | eng |
| Terbitan: |
, 2019
|
| Online Access: |
https://zenodo.org/record/3250598 |
Daftar Isi:
- HPC has evolved in the last years from a technology crucial to the academic research community to a point where it is acknowledged as a key piece of numerical modelling, highlighting different challenges which should be solved. However, these challenges cannot be met by mere extrapolation but require radical innovation in computing technologies and numerical implementations, where the computational evaluation of our models and how to evaluate different alternatives are mandatory. This presentation will provide an overview of the computational performance methodology which could be applied to evaluate the models used by the community and prepare them for the exascale challenge. It will focus mainly on the methodology proposed to evaluate the performance of our atmospheric and ocean models using some profiling tools. The methodologies and tests proposed here could be adapted easily to be applied to other models in order to find main bottlenecks and hotspots for optimizations, including different techniques to evaluate mathematical and algorithimical the model, independently of the hardware restrictions.