Application-driven graph partitioning

Graph partitioning is crucial to parallel computations on large graphs. The choice of partitioning strategies has strong impact on the performance of graph algorithms. For an algorithm of our interest, what partitioning strategy fits it the best and improves its parallel execution? Is it possible to...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:The VLDB journal Ročník 32; číslo 1; s. 149 - 172
Hlavní autori: Fan, Wenfei, Xu, Ruiqi, Yin, Qiang, Yu, Wenyuan, Zhou, Jingren
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Berlin/Heidelberg Springer Berlin Heidelberg 01.01.2023
Springer Nature B.V
Predmet:
Algorithms
Communication
Computer Science
Database Management
Graphs
Partitioning
Performance enhancement
Polynomials
Regular Paper
Workloads
Graph partition
Machine learning
ISSN:1066-8888, 0949-877X
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Graph partitioning is crucial to parallel computations on large graphs. The choice of partitioning strategies has strong impact on the performance of graph algorithms. For an algorithm of our interest, what partitioning strategy fits it the best and improves its parallel execution? Is it possible to provide a uniform partition to a batch of algorithms that run on the same graph simultaneously, and speed up each and every of them? This paper aims to answer these questions. We propose an application-driven hybrid partitioning strategy that, given a graph algorithm A , learns a cost model for A as polynomial regression. We develop partitioners that, given the learned cost model, refine an edge-cut or vertex-cut partition to a hybrid partition and reduce the parallel cost of A . Moreover, we extend the cost-driven strategy to support multiple algorithms at the same time and reduce the parallel cost of each of them. Using real-life and synthetic graphs, we experimentally verify that our partitioning strategy improves the performance of a variety of graph algorithms, up to 22.5 × .
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1066-8888
0949-877X
DOI:10.1007/s00778-022-00736-2