PGX.D a fast distributed graph processing engine

Graph analysis is a powerful method in data analysis. Although several frameworks have been proposed for processing large graph instances in distributed environments, their performance is much lower than using efficient single-machine implementations provided with enough memory. In this paper, we pr...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis S. 1 - 12
Hauptverfasser: Hong, Sungpack, Depner, Siegfried, Manhardt, Thomas, Van Der Lugt, Jan, Verstraaten, Merijn, Chafi, Hassan
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: New York, NY, USA ACM 15.11.2015
Schriftenreihe:ACM Conferences
Schlagworte:
Algorithm design and analysis
Bandwidth
Clustering algorithms
Computational modeling
Computer systems organization > Dependable and fault-tolerant systems and networks
Data models
General and reference > Cross-computing tools and techniques > Performance
Kernel
Mathematics of computing > Discrete mathematics > Graph theory > Graph algorithms
Networks > Network performance evaluation
Programming
Theory of computation > Models of computation > Concurrency
Theory of computation > Models of computation > Concurrency > Parallel computing models
ISBN:1450337236, 9781450337236
ISSN:2167-4337
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Graph analysis is a powerful method in data analysis. Although several frameworks have been proposed for processing large graph instances in distributed environments, their performance is much lower than using efficient single-machine implementations provided with enough memory. In this paper, we present a fast distributed graph processing system, namely PGX.D. We show that PGX.D outperforms other distributed graph systems like GraphLab significantly (3x -- 90x). Furthermore, PGX.D on 4 to 16 machines is also faster than an implementation optimized for single-machine execution. Using a fast cooperative context-switching mechanism, we implement PGX.D as a low-overhead, bandwidth-efficient communication framework that supports remote data-pulling patterns. Moreover, PGX.D achieves large traffic reduction and good workload balance by applying selective ghost nodes, edge partitioning, and edge chunking transparently to the user. Our analysis confirms that each of these features is indeed crucial for overall performance of certain kinds of graph algorithms. Finally, we advocate the use of balanced beefy clusters where the sustained random DRAM-access bandwidth in aggregate is matched with the bandwidth of the underlying interconnection fabric.
ISBN:1450337236
9781450337236
ISSN:2167-4337
DOI:10.1145/2807591.2807620