Understanding computational performance of a complex coupled model like the Energy Exascale Earth System Model (E3SM) poses a singular challenge to domain and computational scientists. Toward that goal, we developed PACE (Performance Analytics for Computational Experiments) , a web-enabled framework to summarize performance data collected from E3SM experiments to derive insights and present them through a web portal.
The primary goal of PACE is to serve as a central hub of performance data to provide an executive summary of E3SM experiments' performance.
https://pace-docs.readthedocs.io
Please cite the PACE software using the following.
DOI: https://doi.org/10.11578/dc.20230509.1
Chicago
Sreepathi, Sarat, KC, Gaurab, Mitchell, Zachary, and Kim, Youngsung. "Performance Analytics for Computational Experiments (PACE)." Computer software. 2023. https://github.com/E3SM-Project/pace. https://doi.org/10.11578/dc.20230509.1.
For your convenience, the following BibTeX entry is provided.
@misc{pace,
title = {{Performance Analytics for Computational Experiments (PACE)}},
author = {Sreepathi, Sarat and KC, Gaurab and Mitchell, Zachary and Kim, Youngsung},
abstractNote = {Performance Analytics for Computational Experiments (PACE) is an end-to-end performance monitoring infrastructure to summarize performance data collected from Department of Energy’s (DOE) climate model experiments to derive insights and present them through a web portal. PACE automatically captures provenance/timing data from simulations run on DOE supercomputers.},
doi = {10.11578/dc.20230509.1},
url = {https://doi.org/10.11578/dc.20230509.1},
howpublished = {[Computer Software] \url{https://doi.org/10.11578/dc.20230509.1}},
year = {2023},
month = {mar}
}