Graph partitioning strategies for efficient BFS in shared-nothing parallel systems

Victor Muntes-Mulero 1 Norbert Martínez-Bazán 1 Josep-Lluís Larriba-Pey 1 Esther Pacitti 2 Patrick Valduriez 2
2 ZENITH - Scientific Data Management
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : Traversing massive graphs as efficiently as possible is essential for many applications. Many common operations on graphs, such as calculating the distance between two nodes, are based on the Breadth First Search traversal. However, because of the exhaustive exploration of all the nodes and edges of the graph, this operation might be very time consuming. A possible solution is distributing the graph among the nodes of a shared-nothing parallel system. Nevertheless, this operation may generate a large amount of inter-node communication. In this paper, we propose two graph partitioning techniques and improve previous distributed versions of BFS in order to reduce this communication.
Document type :
Conference papers
Complete list of metadatas

https://hal-lirmm.ccsd.cnrs.fr/lirmm-00830934
Contributor : Maximilien Servajean <>
Submitted on : Tuesday, April 10, 2018 - 6:38:28 PM
Last modification on : Tuesday, February 19, 2019 - 10:38:45 AM

Links full text

Identifiers

Collections

Citation

Victor Muntes-Mulero, Norbert Martínez-Bazán, Josep-Lluís Larriba-Pey, Esther Pacitti, Patrick Valduriez. Graph partitioning strategies for efficient BFS in shared-nothing parallel systems. WAIM: Web-Age Information Management, Jul 2010, Jiuzhaigou Valley, China. pp.13-24, ⟨10.1007/978-3-642-16720-1_2⟩. ⟨lirmm-00830934⟩

Share

Metrics

Record views

341