StAnD: A Dataset of Linear Static Analysis Problems

TL;DR

This paper introduces StAnD, the largest public dataset of static analysis problems with detailed benchmarks, to facilitate the development and evaluation of sparse linear system solvers in structural engineering.

Contribution

It provides the first extensive public dataset of static analysis problems and benchmarks solver performance on CPU and GPU, aiding research in solver optimization.

Findings

StAnD contains 303,000 static analysis problems.

Benchmark results show differences in solver performance on CPU and GPU.

The dataset and code are publicly available for research use.

Abstract

Static analysis of structures is a fundamental step for determining the stability of structures. Both linear and non-linear static analyses consist of the resolution of sparse linear systems obtained by the finite element method. The development of fast and optimized solvers for sparse linear systems appearing in structural engineering requires data to compare existing approaches, tune algorithms or to evaluate new ideas. We introduce the Static Analysis Dataset (StAnD) containing 303.000 static analysis problems obtained applying realistic loads to simulated frame structures. Along with the dataset, we publish a detailed benchmark comparison of the running time of existing solvers both on CPU and GPU. We release the code used to generate the dataset and benchmark existing solvers on Github. To the best of our knowledge, this is the largest dataset for static analysis problems and it is the first public dataset of sparse linear systems (containing both the matrix and a realistic constant term).