Scaling up data-parallel analytics platforms: Linear algebraic operation cases

Luna Xu; Seung Hwan Lim; Min Li; Ali R. Butt; Ramakrishnan Kannan

doi:10.1109/BigData.2017.8257935

Scaling up data-parallel analytics platforms: Linear algebraic operation cases

Luna Xu, Seung Hwan Lim, Min Li, Ali R. Butt, Ramakrishnan Kannan

Discrete Algorithms

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

Linear algebraic operations such as matrix manipulations form the kernel of many machine learning and other crucial algorithms. Scaling up as well as scaling out such algorithms are key to supporting large scale data analysis that require efficient processing over millions of data samples. To this end, we present, ARION, a hardware acceleration based approach for scaling-up individual tasks of Spark, a popular data-parallel analytics platform. We support both linear algebraic operations of between two dense matrices, and between sparse and dense matrices in distributed environments. ARION provides a flexible control of acceleration according to matrix density, along with efficient scheduling based on runtime resource utilization. We demonstrate the benefit of our approach for general matrix multiplication operations over large matrices with up to four billion elements by using Gramian matrix computation that is commonly used in machine learning. Experiments show that our approach achieves more than 2× and 1.5× end-to-end performance speedups for dense and sparse matrices, respectively, and up to 57.04× faster computation compared to MLlib, a state of the art Spark-based implementation. This work is sponsored in part by the NSF under the grants: CNS-1565314, CNS-1405697, and CNS-1615411. The manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725.

Original language	English
Title of host publication	Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017
Editors	Jian-Yun Nie, Zoran Obradovic, Toyotaro Suzumura, Rumi Ghosh, Raghunath Nambiar, Chonggang Wang, Hui Zang, Ricardo Baeza-Yates, Ricardo Baeza-Yates, Xiaohua Hu, Jeremy Kepner, Alfredo Cuzzocrea, Jian Tang, Masashi Toyoda
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	273-282
Number of pages	10
ISBN (Electronic)	9781538627143
DOIs	https://doi.org/10.1109/BigData.2017.8257935
State	Published - Jul 1 2017
Event	5th IEEE International Conference on Big Data, Big Data 2017 - Boston, United States Duration: Dec 11 2017 → Dec 14 2017

Publication series

Name	Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017
Volume	2018-January

Conference

Conference	5th IEEE International Conference on Big Data, Big Data 2017
Country/Territory	United States
City	Boston
Period	12/11/17 → 12/14/17

Funding

This work is sponsored in part by the NSF under the grants: CNS-1565314, CNS-1405697, and CNS-1615411. The manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725.

Access to Document

10.1109/BigData.2017.8257935

Cite this

Xu, L., Lim, S. H., Li, M., Butt, A. R., & Kannan, R. (2017). Scaling up data-parallel analytics platforms: Linear algebraic operation cases. In J.-Y. Nie, Z. Obradovic, T. Suzumura, R. Ghosh, R. Nambiar, C. Wang, H. Zang, R. Baeza-Yates, R. Baeza-Yates, X. Hu, J. Kepner, A. Cuzzocrea, J. Tang, & M. Toyoda (Eds.), Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017 (pp. 273-282). (Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BigData.2017.8257935

Xu, Luna ; Lim, Seung Hwan ; Li, Min et al. / Scaling up data-parallel analytics platforms : Linear algebraic operation cases. Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017. editor / Jian-Yun Nie ; Zoran Obradovic ; Toyotaro Suzumura ; Rumi Ghosh ; Raghunath Nambiar ; Chonggang Wang ; Hui Zang ; Ricardo Baeza-Yates ; Ricardo Baeza-Yates ; Xiaohua Hu ; Jeremy Kepner ; Alfredo Cuzzocrea ; Jian Tang ; Masashi Toyoda. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 273-282 (Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017).

@inproceedings{d612a22d0b48406fb9dba53b1b10ddca,

title = "Scaling up data-parallel analytics platforms: Linear algebraic operation cases",

abstract = "Linear algebraic operations such as matrix manipulations form the kernel of many machine learning and other crucial algorithms. Scaling up as well as scaling out such algorithms are key to supporting large scale data analysis that require efficient processing over millions of data samples. To this end, we present, ARION, a hardware acceleration based approach for scaling-up individual tasks of Spark, a popular data-parallel analytics platform. We support both linear algebraic operations of between two dense matrices, and between sparse and dense matrices in distributed environments. ARION provides a flexible control of acceleration according to matrix density, along with efficient scheduling based on runtime resource utilization. We demonstrate the benefit of our approach for general matrix multiplication operations over large matrices with up to four billion elements by using Gramian matrix computation that is commonly used in machine learning. Experiments show that our approach achieves more than 2× and 1.5× end-to-end performance speedups for dense and sparse matrices, respectively, and up to 57.04× faster computation compared to MLlib, a state of the art Spark-based implementation. This work is sponsored in part by the NSF under the grants: CNS-1565314, CNS-1405697, and CNS-1615411. The manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725.",

author = "Luna Xu and Lim, {Seung Hwan} and Min Li and Butt, {Ali R.} and Ramakrishnan Kannan",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 5th IEEE International Conference on Big Data, Big Data 2017 ; Conference date: 11-12-2017 Through 14-12-2017",

year = "2017",

month = jul,

day = "1",

doi = "10.1109/BigData.2017.8257935",

language = "English",

series = "Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "273--282",

editor = "Jian-Yun Nie and Zoran Obradovic and Toyotaro Suzumura and Rumi Ghosh and Raghunath Nambiar and Chonggang Wang and Hui Zang and Ricardo Baeza-Yates and Ricardo Baeza-Yates and Xiaohua Hu and Jeremy Kepner and Alfredo Cuzzocrea and Jian Tang and Masashi Toyoda",

booktitle = "Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017",

}

Xu, L, Lim, SH, Li, M, Butt, AR & Kannan, R 2017, Scaling up data-parallel analytics platforms: Linear algebraic operation cases. in J-Y Nie, Z Obradovic, T Suzumura, R Ghosh, R Nambiar, C Wang, H Zang, R Baeza-Yates, R Baeza-Yates, X Hu, J Kepner, A Cuzzocrea, J Tang & M Toyoda (eds), Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017. Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 273-282, 5th IEEE International Conference on Big Data, Big Data 2017, Boston, United States, 12/11/17. https://doi.org/10.1109/BigData.2017.8257935

Scaling up data-parallel analytics platforms: Linear algebraic operation cases. / Xu, Luna; Lim, Seung Hwan; Li, Min et al.
Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017. ed. / Jian-Yun Nie; Zoran Obradovic; Toyotaro Suzumura; Rumi Ghosh; Raghunath Nambiar; Chonggang Wang; Hui Zang; Ricardo Baeza-Yates; Ricardo Baeza-Yates; Xiaohua Hu; Jeremy Kepner; Alfredo Cuzzocrea; Jian Tang; Masashi Toyoda. Institute of Electrical and Electronics Engineers Inc., 2017. p. 273-282 (Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017; Vol. 2018-January).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Scaling up data-parallel analytics platforms

T2 - 5th IEEE International Conference on Big Data, Big Data 2017

AU - Xu, Luna

AU - Lim, Seung Hwan

AU - Li, Min

AU - Butt, Ali R.

AU - Kannan, Ramakrishnan

PY - 2017/7/1

Y1 - 2017/7/1

N2 - Linear algebraic operations such as matrix manipulations form the kernel of many machine learning and other crucial algorithms. Scaling up as well as scaling out such algorithms are key to supporting large scale data analysis that require efficient processing over millions of data samples. To this end, we present, ARION, a hardware acceleration based approach for scaling-up individual tasks of Spark, a popular data-parallel analytics platform. We support both linear algebraic operations of between two dense matrices, and between sparse and dense matrices in distributed environments. ARION provides a flexible control of acceleration according to matrix density, along with efficient scheduling based on runtime resource utilization. We demonstrate the benefit of our approach for general matrix multiplication operations over large matrices with up to four billion elements by using Gramian matrix computation that is commonly used in machine learning. Experiments show that our approach achieves more than 2× and 1.5× end-to-end performance speedups for dense and sparse matrices, respectively, and up to 57.04× faster computation compared to MLlib, a state of the art Spark-based implementation. This work is sponsored in part by the NSF under the grants: CNS-1565314, CNS-1405697, and CNS-1615411. The manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725.

AB - Linear algebraic operations such as matrix manipulations form the kernel of many machine learning and other crucial algorithms. Scaling up as well as scaling out such algorithms are key to supporting large scale data analysis that require efficient processing over millions of data samples. To this end, we present, ARION, a hardware acceleration based approach for scaling-up individual tasks of Spark, a popular data-parallel analytics platform. We support both linear algebraic operations of between two dense matrices, and between sparse and dense matrices in distributed environments. ARION provides a flexible control of acceleration according to matrix density, along with efficient scheduling based on runtime resource utilization. We demonstrate the benefit of our approach for general matrix multiplication operations over large matrices with up to four billion elements by using Gramian matrix computation that is commonly used in machine learning. Experiments show that our approach achieves more than 2× and 1.5× end-to-end performance speedups for dense and sparse matrices, respectively, and up to 57.04× faster computation compared to MLlib, a state of the art Spark-based implementation. This work is sponsored in part by the NSF under the grants: CNS-1565314, CNS-1405697, and CNS-1615411. The manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725.

UR - http://www.scopus.com/inward/record.url?scp=85047783540&partnerID=8YFLogxK

U2 - 10.1109/BigData.2017.8257935

DO - 10.1109/BigData.2017.8257935

M3 - Conference contribution

AN - SCOPUS:85047783540

T3 - Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017

SP - 273

EP - 282

BT - Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017

A2 - Nie, Jian-Yun

A2 - Obradovic, Zoran

A2 - Suzumura, Toyotaro

A2 - Ghosh, Rumi

A2 - Nambiar, Raghunath

A2 - Wang, Chonggang

A2 - Zang, Hui

A2 - Baeza-Yates, Ricardo

A2 - Hu, Xiaohua

A2 - Kepner, Jeremy

A2 - Cuzzocrea, Alfredo

A2 - Tang, Jian

A2 - Toyoda, Masashi

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 11 December 2017 through 14 December 2017

ER -

Xu L, Lim SH, Li M, Butt AR, Kannan R. Scaling up data-parallel analytics platforms: Linear algebraic operation cases. In Nie JY, Obradovic Z, Suzumura T, Ghosh R, Nambiar R, Wang C, Zang H, Baeza-Yates R, Baeza-Yates R, Hu X, Kepner J, Cuzzocrea A, Tang J, Toyoda M, editors, Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 273-282. (Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017). doi: 10.1109/BigData.2017.8257935

Scaling up data-parallel analytics platforms: Linear algebraic operation cases

Abstract

Publication series

Conference

Funding

Access to Document

Other files and links

Fingerprint

Cite this