Parallel Neighborhood Expansion Based Graph Edge Partitioning Algorithm

This paper proposes a novel parallel neighborhood expansion-based algorithm for graph edge partitioning aimed at addressing computational efficiency and scalability issues in large-scale graph data processing. The algorithm employs innovative parallel strategies and load balancing techniques, effect...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE International Conference on Power, Intelligent Computing and Systems (Online) s. 1583 - 1587
Hlavní autoři: Huang, Jijie, Chen, Yang
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 26.07.2024
Témata:
Computational efficiency
Data processing
Distance measurement
graph edge partitioning
high-performance computing
large-scale graph processing
load balancing
Load management
Memory management
parallel algorithm
Partitioning algorithms
Scalability
Social networking (online)
ISSN:2834-8567
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper proposes a novel parallel neighborhood expansion-based algorithm for graph edge partitioning aimed at addressing computational efficiency and scalability issues in large-scale graph data processing. The algorithm employs innovative parallel strategies and load balancing techniques, effectively enhancing the speed and quality of graph edge partitioning. Experimental results demonstrate outstanding performance of the algorithm when handling graphs ranging one hundred thousand to millions of edges. Compared to existing methods, it significantly improves execution speed while maintaining comparable partitioning quality. In highly parallel environments, the algorithm exhibits excellent scalability and load balancing performance. Moreover, it demonstrates impressive memory efficiency, enabling processing of even larger-scale graph data.
ISSN:2834-8567
DOI:10.1109/ICPICS62053.2024.10796853