Sorting big data on heterogeneous near-data processing systems

Erik Vermij; Leandro Fiorin; Christoph Hagleitner; Koen Bertels

doi:10.1145/3075564.3078885

Sorting big data on heterogeneous near-data processing systems

Erik Vermij, Leandro Fiorin, Christoph Hagleitner, Koen Bertels

Quantum & Computer Engineering

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

4 Citations (Scopus)

Abstract

Big data workloads assumed recently a relevant importance in many business and scientific applications. Sorting ele-ments efficiently in big data workloads is a key operation. In this work, we analyze the implementation of the mergesort algorithm on heterogeneous systems composed of CPUs and near-data processors located on the system memory channels. For configurations with equal number of active CPU cores and near-data processors, our experiments show a per-formance speedup of up to 2.5, as well as up to 2.5× energy-per-solution reduction.

Original language	English
Title of host publication	CF'17 Proceedings of the Computing Frontiers Conference
Place of Publication	New York
Publisher	Association for Computing Machinery (ACM)
Pages	349-354
Number of pages	6
ISBN (Print)	978-1-4503-4487-6
DOIs	https://doi.org/10.1145/3075564.3078885
Publication status	Published - 2017
Event	ACM International Conference on Computing Frontiers 2017: CF'17 - Siena, Italy Duration: 15 May 2017 → 17 May 2017

Conference

Conference	ACM International Conference on Computing Frontiers 2017
Country/Territory	Italy
City	Siena
Period	15/05/17 → 17/05/17

Access to Document

10.1145/3075564.3078885

Cite this

@inproceedings{794fa91ec76c4e6791429e8d9ef5553a,

title = "Sorting big data on heterogeneous near-data processing systems",

abstract = "Big data workloads assumed recently a relevant importance in many business and scientific applications. Sorting ele-ments efficiently in big data workloads is a key operation. In this work, we analyze the implementation of the mergesort algorithm on heterogeneous systems composed of CPUs and near-data processors located on the system memory channels. For configurations with equal number of active CPU cores and near-data processors, our experiments show a per-formance speedup of up to 2.5, as well as up to 2.5× energy-per-solution reduction.",

author = "Erik Vermij and Leandro Fiorin and Christoph Hagleitner and Koen Bertels",

year = "2017",

doi = "10.1145/3075564.3078885",

language = "English",

isbn = "978-1-4503-4487-6",

pages = "349--354",

booktitle = "CF'17 Proceedings of the Computing Frontiers Conference",

publisher = "Association for Computing Machinery (ACM)",

address = "United States",

note = "ACM International Conference on Computing Frontiers 2017 : CF'17 ; Conference date: 15-05-2017 Through 17-05-2017",

}

Sorting big data on heterogeneous near-data processing systems. / Vermij, Erik; Fiorin, Leandro; Hagleitner, Christoph et al.
CF'17 Proceedings of the Computing Frontiers Conference . New York: Association for Computing Machinery (ACM), 2017. p. 349-354.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Sorting big data on heterogeneous near-data processing systems

AU - Vermij, Erik

AU - Fiorin, Leandro

AU - Hagleitner, Christoph

AU - Bertels, Koen

PY - 2017

Y1 - 2017

N2 - Big data workloads assumed recently a relevant importance in many business and scientific applications. Sorting ele-ments efficiently in big data workloads is a key operation. In this work, we analyze the implementation of the mergesort algorithm on heterogeneous systems composed of CPUs and near-data processors located on the system memory channels. For configurations with equal number of active CPU cores and near-data processors, our experiments show a per-formance speedup of up to 2.5, as well as up to 2.5× energy-per-solution reduction.

AB - Big data workloads assumed recently a relevant importance in many business and scientific applications. Sorting ele-ments efficiently in big data workloads is a key operation. In this work, we analyze the implementation of the mergesort algorithm on heterogeneous systems composed of CPUs and near-data processors located on the system memory channels. For configurations with equal number of active CPU cores and near-data processors, our experiments show a per-formance speedup of up to 2.5, as well as up to 2.5× energy-per-solution reduction.

UR - http://www.scopus.com/inward/record.url?scp=85027043270&partnerID=8YFLogxK

U2 - 10.1145/3075564.3078885

DO - 10.1145/3075564.3078885

M3 - Conference contribution

AN - SCOPUS:85027043270

SN - 978-1-4503-4487-6

SP - 349

EP - 354

BT - CF'17 Proceedings of the Computing Frontiers Conference

PB - Association for Computing Machinery (ACM)

CY - New York

T2 - ACM International Conference on Computing Frontiers 2017

Y2 - 15 May 2017 through 17 May 2017

ER -

Sorting big data on heterogeneous near-data processing systems

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this