Task-based parallel breadth-first search in heterogeneous environments

Lluis Miquel Munguia, David A. Bader, Eduard Ayguade

Research output: Chapter in Book/Report/Conference proceedingConference contribution

13 Scopus citations

Abstract

Breadth-first search (BFS) is an essential graph traversal strategy widely used in many computing applications. Because of its irregular data access patterns, BFS has become a non-trivial problem hard to parallelize efficiently. In this paper, we introduce a parallelization strategy that allows the load balancing of computation resources as well as the execution of graph traversals in hybrid environments composed of CPUs and GPUs. To achieve that goal, we use a fine-grained task-based parallelization scheme and the OmpSs programming model. We obtain processing rates up to 2.8 billion traversed edges per second with a single GPU and a multi-core processor. Our study shows high processing rates are achievable with hybrid environments despite the GPU communication latency and memory coherence.

Original languageEnglish (US)
Title of host publication2012 19th International Conference on High Performance Computing, HiPC 2012
DOIs
StatePublished - 2012
Externally publishedYes
Event2012 19th International Conference on High Performance Computing, HiPC 2012 - Pune, India
Duration: Dec 18 2012Dec 21 2012

Publication series

Name2012 19th International Conference on High Performance Computing, HiPC 2012

Conference

Conference2012 19th International Conference on High Performance Computing, HiPC 2012
Country/TerritoryIndia
CityPune
Period12/18/1212/21/12

All Science Journal Classification (ASJC) codes

  • Software

Fingerprint

Dive into the research topics of 'Task-based parallel breadth-first search in heterogeneous environments'. Together they form a unique fingerprint.

Cite this