Improving the efficiency of graph algorithm executions on high‐performance computing

Summary The growing need for extracting information from large graphs has been pushing the development of parallel graph algorithms. However, the highly irregular structure of the real‐world graphs limits the performance and energy improvements of graph applications. In this paper, we show that, in...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Concurrency and computation Jg. 35; H. 18
Hauptverfasser: Moori, Marcelo K., A. Rocha, Hiago Mayk G., Schwarzrock, Janaina, Lorenzon, Arthur F., Beck, Antonio Carlos S.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Hoboken Wiley Subscription Services, Inc 15.08.2023
Schlagworte:
Algorithms
Efficiency
Energy consumption
energy‐saving
graph analytics
Graphs
High performance computing
Microprocessors
Multiprocessing
multiprogramming
parallel graph processing
system efficiency
ISSN:1532-0626, 1532-0634
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Summary The growing need for extracting information from large graphs has been pushing the development of parallel graph algorithms. However, the highly irregular structure of the real‐world graphs limits the performance and energy improvements of graph applications. In this paper, we show that, in most cases, using all the available cores of the multiprocessor is not the best option in terms of the aforementioned non‐functional requirements. Based on that, we propose GraphKat, a framework that enables the simultaneous processing of several algorithms/graphs instead of executing them serially (i.e., one after another), increasing efficiency in terms of performance and energy. GraphKat works in two steps: (i) it characterizes the graph applications with a specific number of threads based on their efficiency levels; and (ii) it defines the execution order of all graph applications in the target system. Experimental results on three multicore processors (Intel and AMD) show that GraphKat improves the overall system's efficiency related to performance (up to 434.26×$$ 434.26\times $$) and energy‐saving (up to 245.21×$$ \times $$), and reduces the graph applications' execution time (up to 17.70×$$ 17.70\times $$) and energy consumption (up to 6.64×$$ \times $$) compared to the default execution of parallel applications on HPC systems.
Bibliographie:Funding information
Conselho Nacional de Desenvolvimento Científico e Tecnológico, Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, Grant/Award Number: 001; Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1532-0626
1532-0634
DOI:10.1002/cpe.7419