Energy efficiency of large scale graph processing platforms

Kashif Nizam Khan, Mohammad Ashraful Hoque, Tapio Niemi, Zhonghong Ou, Jukka K. Nurminen

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

3 Citations (Scopus)

Abstract

A number of graph processing platforms have emerged recently as a result of the growing demand on graph data analytics with complex and large-scale graph structured datasets. These platforms have been tailored for iterative graph computations and can offer an order of magnitude performance gain over generic data-flow frameworks like Apache Hadoop and Spark. Nevertheless, the increasing availability of such platforms and their functionality overlap necessitates a comparative study on various aspects of the platforms, including applications, performance and energy efficiency. In this work, we focus on the energy efficiency aspect of some large scale graph processing platforms. Specifically, we select two representatives, e.g., Apache Giraph and Spark GraphX, for the comparative study. We compare and analyze the energy consumption of these two platforms with PageRank, Strongly Connected Component and Single Source Shortest Path algorithms over five different realistic graphs. Our experimental results demonstrate that GraphX outperforms Giraph in terms of energy consumption. Specifically, Giraph consumes 1.71 times more energy than GraphX on average for the mentioned algorithms.
Original languageEnglish
Title of host publicationProceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct - UbiComp '16
Place of PublicationNew York
PublisherAssociation for Computing Machinery ACM
Pages1287-1294
Number of pages8
ISBN (Print)978-1-4503-4462-3
DOIs
Publication statusPublished - 2016
MoE publication typeA4 Article in a conference publication

Fingerprint

Electric sparks
Energy efficiency
Energy utilization
Processing
Availability

Cite this

Khan, K. N., Hoque, M. A., Niemi, T., Ou, Z., & Nurminen, J. K. (2016). Energy efficiency of large scale graph processing platforms. In Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct - UbiComp '16 (pp. 1287-1294). New York: Association for Computing Machinery ACM. https://doi.org/10.1145/2968219.2968296
Khan, Kashif Nizam ; Hoque, Mohammad Ashraful ; Niemi, Tapio ; Ou, Zhonghong ; Nurminen, Jukka K. / Energy efficiency of large scale graph processing platforms. Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct - UbiComp '16. New York : Association for Computing Machinery ACM, 2016. pp. 1287-1294
@inproceedings{87bfabd9356644698b0608f7f0f04cd9,
title = "Energy efficiency of large scale graph processing platforms",
abstract = "A number of graph processing platforms have emerged recently as a result of the growing demand on graph data analytics with complex and large-scale graph structured datasets. These platforms have been tailored for iterative graph computations and can offer an order of magnitude performance gain over generic data-flow frameworks like Apache Hadoop and Spark. Nevertheless, the increasing availability of such platforms and their functionality overlap necessitates a comparative study on various aspects of the platforms, including applications, performance and energy efficiency. In this work, we focus on the energy efficiency aspect of some large scale graph processing platforms. Specifically, we select two representatives, e.g., Apache Giraph and Spark GraphX, for the comparative study. We compare and analyze the energy consumption of these two platforms with PageRank, Strongly Connected Component and Single Source Shortest Path algorithms over five different realistic graphs. Our experimental results demonstrate that GraphX outperforms Giraph in terms of energy consumption. Specifically, Giraph consumes 1.71 times more energy than GraphX on average for the mentioned algorithms.",
author = "Khan, {Kashif Nizam} and Hoque, {Mohammad Ashraful} and Tapio Niemi and Zhonghong Ou and Nurminen, {Jukka K.}",
year = "2016",
doi = "10.1145/2968219.2968296",
language = "English",
isbn = "978-1-4503-4462-3",
pages = "1287--1294",
booktitle = "Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct - UbiComp '16",
publisher = "Association for Computing Machinery ACM",
address = "United States",

}

Khan, KN, Hoque, MA, Niemi, T, Ou, Z & Nurminen, JK 2016, Energy efficiency of large scale graph processing platforms. in Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct - UbiComp '16. Association for Computing Machinery ACM, New York, pp. 1287-1294. https://doi.org/10.1145/2968219.2968296

Energy efficiency of large scale graph processing platforms. / Khan, Kashif Nizam; Hoque, Mohammad Ashraful; Niemi, Tapio; Ou, Zhonghong; Nurminen, Jukka K.

Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct - UbiComp '16. New York : Association for Computing Machinery ACM, 2016. p. 1287-1294.

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - Energy efficiency of large scale graph processing platforms

AU - Khan, Kashif Nizam

AU - Hoque, Mohammad Ashraful

AU - Niemi, Tapio

AU - Ou, Zhonghong

AU - Nurminen, Jukka K.

PY - 2016

Y1 - 2016

N2 - A number of graph processing platforms have emerged recently as a result of the growing demand on graph data analytics with complex and large-scale graph structured datasets. These platforms have been tailored for iterative graph computations and can offer an order of magnitude performance gain over generic data-flow frameworks like Apache Hadoop and Spark. Nevertheless, the increasing availability of such platforms and their functionality overlap necessitates a comparative study on various aspects of the platforms, including applications, performance and energy efficiency. In this work, we focus on the energy efficiency aspect of some large scale graph processing platforms. Specifically, we select two representatives, e.g., Apache Giraph and Spark GraphX, for the comparative study. We compare and analyze the energy consumption of these two platforms with PageRank, Strongly Connected Component and Single Source Shortest Path algorithms over five different realistic graphs. Our experimental results demonstrate that GraphX outperforms Giraph in terms of energy consumption. Specifically, Giraph consumes 1.71 times more energy than GraphX on average for the mentioned algorithms.

AB - A number of graph processing platforms have emerged recently as a result of the growing demand on graph data analytics with complex and large-scale graph structured datasets. These platforms have been tailored for iterative graph computations and can offer an order of magnitude performance gain over generic data-flow frameworks like Apache Hadoop and Spark. Nevertheless, the increasing availability of such platforms and their functionality overlap necessitates a comparative study on various aspects of the platforms, including applications, performance and energy efficiency. In this work, we focus on the energy efficiency aspect of some large scale graph processing platforms. Specifically, we select two representatives, e.g., Apache Giraph and Spark GraphX, for the comparative study. We compare and analyze the energy consumption of these two platforms with PageRank, Strongly Connected Component and Single Source Shortest Path algorithms over five different realistic graphs. Our experimental results demonstrate that GraphX outperforms Giraph in terms of energy consumption. Specifically, Giraph consumes 1.71 times more energy than GraphX on average for the mentioned algorithms.

U2 - 10.1145/2968219.2968296

DO - 10.1145/2968219.2968296

M3 - Conference article in proceedings

SN - 978-1-4503-4462-3

SP - 1287

EP - 1294

BT - Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct - UbiComp '16

PB - Association for Computing Machinery ACM

CY - New York

ER -

Khan KN, Hoque MA, Niemi T, Ou Z, Nurminen JK. Energy efficiency of large scale graph processing platforms. In Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct - UbiComp '16. New York: Association for Computing Machinery ACM. 2016. p. 1287-1294 https://doi.org/10.1145/2968219.2968296