Bin-VBSR: Variable Block Size Binned Block-Compressed Sparse Row for Efficient GPU-Accelerated Finite Element Analysis
| dc.contributor.author | Pfeil, Florian | en_US |
| dc.contributor.author | Ferreira, Stephanie | en_US |
| dc.contributor.author | Mueller-Roemer, Johannes Sebastian | en_US |
| dc.contributor.editor | Egger, Bernhard | en_US |
| dc.contributor.editor | Günther, Tobias | en_US |
| dc.date.accessioned | 2025-09-24T10:38:28Z | |
| dc.date.available | 2025-09-24T10:38:28Z | |
| dc.date.issued | 2025 | |
| dc.description.abstract | We present Binned Variable Block Compressed Sparse Row (Bin-VBSR), a novel GPU-optimized sparse matrix data structure and associated sparse matrix-vector multiplication algorithm for matrices with variable-size dense blocks. This includes a novel approach to handling long rows in the Binned Compressed Sparse Row (Bin-CSR) family of GPU-optimized sparse matrix data structures. We demonstrate speedups of up to 9.9× over Bin-BCSR* and extend its data compression advantages over compressed sparse row (CSR) to variable block size, resulting in an improvement of up to 50%. | en_US |
| dc.description.sectionheaders | Geometry, Simulation, and Optimization | |
| dc.description.seriesinformation | Vision, Modeling, and Visualization | |
| dc.identifier.doi | 10.2312/vmv.20251245 | |
| dc.identifier.isbn | 978-3-03868-294-3 | |
| dc.identifier.pages | 8 pages | |
| dc.identifier.uri | https://doi.org/10.2312/vmv.20251245 | |
| dc.identifier.uri | https://diglib.eg.org/handle/10.2312/vmv20251245 | |
| dc.publisher | The Eurographics Association | en_US |
| dc.rights | Attribution 4.0 International License | |
| dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | |
| dc.subject | CCS Concepts: Computing methodologies → Massively parallel algorithms; Mathematics of computing → Computations on matrices; Mathematical software performance | |
| dc.subject | Computing methodologies → Massively parallel algorithms | |
| dc.subject | Mathematics of computing → Computations on matrices | |
| dc.subject | Mathematical software performance | |
| dc.title | Bin-VBSR: Variable Block Size Binned Block-Compressed Sparse Row for Efficient GPU-Accelerated Finite Element Analysis | en_US |