A Non-linear GPU Thread Map for Triangular Domains

Cristobal A Navarro, Benjamín Bustos, Nancy Hitschfeld Kahler
September 2016
arXiv

Abstract

We study the case of mapping threads efficiently onto triangular domain problems and propose a block-space linear map based on the properties of the lower triangular matrix that reduces the number of unnecessary threads from O(n^2) to O(n). Performance results for global memory accesses show an improvement of up to 18% with respect to the bounding-box approach. For shared memory scenarios the proposed map was the fastest approach achieving 7% of performance improvement while preserving thread locality.

Type
Preprint
Publication
arXiv
GPU Computing Thread Mapping Triangular Domains
Cristobal A Navarro
Cristobal A Navarro
Professor at the Universidad Austral de Chile

Professor at the Universidad Austral de Chile

Nancy Hitschfeld Kahler
Nancy Hitschfeld Kahler
+Lab founder | Full Professor Universidad de Chile

Full Professor at the Department of Computer Science, University of Chile. Her main research interests include geometric modeling, geometric meshes, and parallel algorithms (GPU computing), focused in computational science, and engineering applications.