GraphRay

Distributed pathfinder network scaling

Alessio Arleo, Oh Hyun Kwon, Kwan-Liu Ma

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Pathfinder network scaling is a graph sparsification technique that has been popularly used due to its efficacy of extracting the 'important' structure of a graph. However, existing algorithms to compute the pathfinder network (PFNET) of a graph have prohibitively expensive time complexity for large graphs: O(n3) for the general case and O(n2 log n) for a specific parameter setting, PFNET(r = ∞, q = n - 1), which is considered in many applications. In this paper, we introduce the first distributed technique to compute the pathfinder network with the specific parameters (r = ∞ and q = n - 1) of a large graph with millions of edges. The results of our experiments show our technique is scalable; it efficiently utilizes a parallel distributed computing environment, reducing the running times as more processing units are added.

Original languageEnglish (US)
Title of host publication2017 IEEE 7th Symposium on Large Data Analysis and Visualization, LDAV 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages74-83
Number of pages10
Volume2017-December
ISBN (Electronic)9781538606179
DOIs
StatePublished - Dec 19 2017
Event7th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2017 - Phoenix, United States
Duration: Oct 2 2017 → …

Other

Other7th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2017
CountryUnited States
CityPhoenix
Period10/2/17 → …

Fingerprint

Distributed computer systems
scaling
Processing
Experiments
experiment
time

ASJC Scopus subject areas

  • Hardware and Architecture
  • Information Systems
  • Media Technology
  • Library and Information Sciences

Cite this

Arleo, A., Kwon, O. H., & Ma, K-L. (2017). GraphRay: Distributed pathfinder network scaling. In 2017 IEEE 7th Symposium on Large Data Analysis and Visualization, LDAV 2017 (Vol. 2017-December, pp. 74-83). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/LDAV.2017.8231853

GraphRay : Distributed pathfinder network scaling. / Arleo, Alessio; Kwon, Oh Hyun; Ma, Kwan-Liu.

2017 IEEE 7th Symposium on Large Data Analysis and Visualization, LDAV 2017. Vol. 2017-December Institute of Electrical and Electronics Engineers Inc., 2017. p. 74-83.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Arleo, A, Kwon, OH & Ma, K-L 2017, GraphRay: Distributed pathfinder network scaling. in 2017 IEEE 7th Symposium on Large Data Analysis and Visualization, LDAV 2017. vol. 2017-December, Institute of Electrical and Electronics Engineers Inc., pp. 74-83, 7th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2017, Phoenix, United States, 10/2/17. https://doi.org/10.1109/LDAV.2017.8231853
Arleo A, Kwon OH, Ma K-L. GraphRay: Distributed pathfinder network scaling. In 2017 IEEE 7th Symposium on Large Data Analysis and Visualization, LDAV 2017. Vol. 2017-December. Institute of Electrical and Electronics Engineers Inc. 2017. p. 74-83 https://doi.org/10.1109/LDAV.2017.8231853
Arleo, Alessio ; Kwon, Oh Hyun ; Ma, Kwan-Liu. / GraphRay : Distributed pathfinder network scaling. 2017 IEEE 7th Symposium on Large Data Analysis and Visualization, LDAV 2017. Vol. 2017-December Institute of Electrical and Electronics Engineers Inc., 2017. pp. 74-83
@inproceedings{cb431428b56247838cb19f0601daa785,
title = "GraphRay: Distributed pathfinder network scaling",
abstract = "Pathfinder network scaling is a graph sparsification technique that has been popularly used due to its efficacy of extracting the 'important' structure of a graph. However, existing algorithms to compute the pathfinder network (PFNET) of a graph have prohibitively expensive time complexity for large graphs: O(n3) for the general case and O(n2 log n) for a specific parameter setting, PFNET(r = ∞, q = n - 1), which is considered in many applications. In this paper, we introduce the first distributed technique to compute the pathfinder network with the specific parameters (r = ∞ and q = n - 1) of a large graph with millions of edges. The results of our experiments show our technique is scalable; it efficiently utilizes a parallel distributed computing environment, reducing the running times as more processing units are added.",
author = "Alessio Arleo and Kwon, {Oh Hyun} and Kwan-Liu Ma",
year = "2017",
month = "12",
day = "19",
doi = "10.1109/LDAV.2017.8231853",
language = "English (US)",
volume = "2017-December",
pages = "74--83",
booktitle = "2017 IEEE 7th Symposium on Large Data Analysis and Visualization, LDAV 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - GraphRay

T2 - Distributed pathfinder network scaling

AU - Arleo, Alessio

AU - Kwon, Oh Hyun

AU - Ma, Kwan-Liu

PY - 2017/12/19

Y1 - 2017/12/19

N2 - Pathfinder network scaling is a graph sparsification technique that has been popularly used due to its efficacy of extracting the 'important' structure of a graph. However, existing algorithms to compute the pathfinder network (PFNET) of a graph have prohibitively expensive time complexity for large graphs: O(n3) for the general case and O(n2 log n) for a specific parameter setting, PFNET(r = ∞, q = n - 1), which is considered in many applications. In this paper, we introduce the first distributed technique to compute the pathfinder network with the specific parameters (r = ∞ and q = n - 1) of a large graph with millions of edges. The results of our experiments show our technique is scalable; it efficiently utilizes a parallel distributed computing environment, reducing the running times as more processing units are added.

AB - Pathfinder network scaling is a graph sparsification technique that has been popularly used due to its efficacy of extracting the 'important' structure of a graph. However, existing algorithms to compute the pathfinder network (PFNET) of a graph have prohibitively expensive time complexity for large graphs: O(n3) for the general case and O(n2 log n) for a specific parameter setting, PFNET(r = ∞, q = n - 1), which is considered in many applications. In this paper, we introduce the first distributed technique to compute the pathfinder network with the specific parameters (r = ∞ and q = n - 1) of a large graph with millions of edges. The results of our experiments show our technique is scalable; it efficiently utilizes a parallel distributed computing environment, reducing the running times as more processing units are added.

UR - http://www.scopus.com/inward/record.url?scp=85048406744&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048406744&partnerID=8YFLogxK

U2 - 10.1109/LDAV.2017.8231853

DO - 10.1109/LDAV.2017.8231853

M3 - Conference contribution

VL - 2017-December

SP - 74

EP - 83

BT - 2017 IEEE 7th Symposium on Large Data Analysis and Visualization, LDAV 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -