Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Efficient Stream Compaction on Wide SIMD Many-Core Architectures Markus Billeter

Summary: Efficient Stream Compaction on Wide SIMD Many-Core Architectures
Markus Billeter
Chalmers University of Technology
Ola Olsson
Chalmers University of Technology
Ulf Assarsson
Chalmers University of Technology
Stream compaction is a common parallel primitive used to remove
unwanted elements in sparse data. This allows highly parallel algo-
rithms to maintain performance over several processing steps and
reduces overall memory usage.
For wide SIMD many-core architectures, we present a novel stream
compaction algorithm and explore several variations thereof. Our
algorithm is designed to maximize concurrent execution, with min-
imal use of synchronization. Bandwidth and auxiliary storage re-
quirements are reduced significantly, which allows for substantially
better performance.
We have tested our algorithms using CUDA on a PC with an
NVIDIA GeForce GTX280 GPU. On this hardware, our reference


Source: Assarsson, Ulf - Department of Computer Engineering, Chalmers University of Technology


Collections: Computer Technologies and Information Sciences