UPDF AI

Efficient stream compaction on wide SIMD many-core architectures

Markus Billeter,Ola Olsson,Ulf Assarsson

2009 · DOI: 10.1145/1572769.1572795
High Performance Graphics · 124 Citations

TLDR

This work presents a novel stream compaction algorithm designed to maximize concurrent execution, with minimal use of synchronization, for wide SIMD many-core architectures and explores several variations thereof.