An effective GPU implementation of breadth-first search

Breadth-first search (BFS) has wide applications in electronic design automation (EDA) as well as in other fields. Researchers have tried to accelerate BFS on the GPU, but the two published works are both asymptotically slower than the fastest CPU implementation. In this paper, we present a new GPU...

Full description

Saved in:
Bibliographic Details
Published in:Design Automation Conference pp. 52 - 55
Main Authors: Lijuan Luo, Wong, Martin, Wen-mei Hwu
Format: Conference Proceeding
Language:English
Published: IEEE 13.06.2010
Subjects:
Acceleration
Application software
BFS
Central Processing Unit
Compaction
Computational complexity
CUDA
Electronic design automation and methodology
GPU computing
Kernel
Mathematical programming
Mathematics
Permission
ISBN:9781424466771, 1424466776
ISSN:0738-100X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Breadth-first search (BFS) has wide applications in electronic design automation (EDA) as well as in other fields. Researchers have tried to accelerate BFS on the GPU, but the two published works are both asymptotically slower than the fastest CPU implementation. In this paper, we present a new GPU implementation of BFS that uses a hierarchical queue management technique and a three-layer kernel arrangement strategy. It guarantees the same computational complexity as the fastest sequential version and can achieve up to 10 times speedup.
ISBN:9781424466771
1424466776
ISSN:0738-100X
DOI:10.1145/1837274.1837289