Linear Algebra Aware Benchmarks (LAAB) is a framework for systematically assessing and reporting the performance of mathematical library installations on HPC systems. Mathematical libraries provide interfaces for operations that form the computational building blocks of scientific applications. Reporting their performance is important for assessing application efficiency, estimating compute-time requirements, and preparing resource-allocation requests.
This project contains the software components of the LAAB framework and serves as a supplement to the ongoing PhD work, "Methods for Performance Reporting of Mathematical Library Installations on HPC Systems" at RWTH Aachen University, Germany. The framework was developed to aid HPC user-support activities at the Juelich Supercomputing Centre, Forschungszentrum Juelich, Germany and, more broadly, at HPC centres in general.
LAAB-HPC was developed with the support of Juelich Supercomputing Center at Forschungszentrum Juelich, Germany through the BMBF project 01-1H1-6013 AP6-NRW Anwenderunterstuetzung SiVeGCS, the eSSENCE Programme under the Swedish Government’s Strategic Research Initiative, and RWTH Aachen University, Germany through the DFG project IRTG-2379.
LAAB-HPC has several software components. Unless otherwise specified by an individual component, the software components are distributed under the GNU AGPL-3.0 License. Components derived from or incorporating third-party software retain their respective copyright notices and licence terms.
Copyright (c) 2026 Forschungszentrum Juelich GmbH, Juelich Supercomputing Centre
Citations:
Preprint: Sankaran, A., and P. Bientinesi. Performance Reporting of Mathematical Library Installations with LAAB – An Overview. arXiv, 13 Aug. 2026, https://doi.org/10.48550/arXiv.2608.13512
Code: Sankaran, A. LAAB-HPC. Version 2026a, Zenodo, 13 Aug. 2026, https://doi.org/10.5281/zenodo.21921183.
