Joint Schedule and Layout Autotuning for Sparse Matrices with Compound Entries on GPUs
Loading...
Date
2019
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association
Abstract
Large sparse matrices with compound entries, i.e., complex and quaternionic matrices as well as matrices with dense blocks, are a core component of many algorithms in geometry processing, physically based animation, and other areas of computer graphics. We generalize several matrix layouts and apply joint schedule and layout autotuning to improve the performance of the sparse matrix-vector product on massively parallel graphics processing units. Compared to schedule tuning without layout tuning, we achieve speedups of up to 5.5x. In comparison to cuSPARSE, we achieve speedups of up to 4.7x
Description
@inproceedings{10.2312:vmv.20191324,
booktitle = {Vision, Modeling and Visualization},
editor = {Schulz, Hans-Jörg and Teschner, Matthias and Wimmer, Michael},
title = {{Joint Schedule and Layout Autotuning for Sparse Matrices with Compound Entries on GPUs}},
author = {Mueller-Roemer, Johannes Sebastian and Stork, André and Fellner, Dieter W.},
year = {2019},
publisher = {The Eurographics Association},
ISBN = {978-3-03868-098-7},
DOI = {10.2312/vmv.20191324}
}