DPM: Data Partitioning Method for pipelined MapReduce on GPU

Myung Hyun Jo; Won Woo Ro

doi:10.1109/ISCE.2014.6884382

DPM: Data Partitioning Method for pipelined MapReduce on GPU

Myung Hyun Jo, Won Woo Ro

Department of Electrical and Electronic Engineering

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

Abstract

The MapReduce frameworks using a modern graphic processor (GPU) have improved the performance of data-intensive applications. While the prior researches have enhanced the parallelism of the MapReduce application on a GPU, archiving optimal distribution of big data on heterogeneous devices is still a challengeable issue. We therefore propose a method to evenly separate the computing cost under limited memory size. To solve this problem, we design and propose DPM, a Data Partitioning Method, using a GPU to smartly distribute workload of MapReduce. The proposed technique provides well-balanced processing cost for heterogeneous devices.

Original language	English
Title of host publication	ISCE 2014 - 18th IEEE International Symposium on Consumer Electronics
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Print)	9781479945924
DOIs	https://doi.org/10.1109/ISCE.2014.6884382
Publication status	Published - 2014
Event	18th IEEE International Symposium on Consumer Electronics, ISCE 2014 - Jeju, Korea, Republic of Duration: 2014 Jun 22 → 2014 Jun 25

Publication series

Name	Proceedings of the International Symposium on Consumer Electronics, ISCE

Other

Other	18th IEEE International Symposium on Consumer Electronics, ISCE 2014
Country	Korea, Republic of
City	Jeju
Period	14/6/22 → 14/6/25

All Science Journal Classification (ASJC) codes

Engineering(all)

Access to Document

10.1109/ISCE.2014.6884382

Fingerprint Dive into the research topics of 'DPM: Data Partitioning Method for pipelined MapReduce on GPU'. Together they form a unique fingerprint.

Cite this

@inproceedings{57979f5216284110b872088163c45bbf,

title = "DPM: Data Partitioning Method for pipelined MapReduce on GPU",

abstract = "The MapReduce frameworks using a modern graphic processor (GPU) have improved the performance of data-intensive applications. While the prior researches have enhanced the parallelism of the MapReduce application on a GPU, archiving optimal distribution of big data on heterogeneous devices is still a challengeable issue. We therefore propose a method to evenly separate the computing cost under limited memory size. To solve this problem, we design and propose DPM, a Data Partitioning Method, using a GPU to smartly distribute workload of MapReduce. The proposed technique provides well-balanced processing cost for heterogeneous devices.",

author = "Jo, {Myung Hyun} and Ro, {Won Woo}",

year = "2014",

doi = "10.1109/ISCE.2014.6884382",

language = "English",

isbn = "9781479945924",

series = "Proceedings of the International Symposium on Consumer Electronics, ISCE",

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

booktitle = "ISCE 2014 - 18th IEEE International Symposium on Consumer Electronics",

address = "United States",

}

Jo, MH & Ro, WW 2014, DPM: Data Partitioning Method for pipelined MapReduce on GPU. in ISCE 2014 - 18th IEEE International Symposium on Consumer Electronics., 6884382, Proceedings of the International Symposium on Consumer Electronics, ISCE, Institute of Electrical and Electronics Engineers Inc., 18th IEEE International Symposium on Consumer Electronics, ISCE 2014, Jeju, Korea, Republic of, 14/6/22. https://doi.org/10.1109/ISCE.2014.6884382

DPM : Data Partitioning Method for pipelined MapReduce on GPU. / Jo, Myung Hyun; Ro, Won Woo.

ISCE 2014 - 18th IEEE International Symposium on Consumer Electronics. Institute of Electrical and Electronics Engineers Inc., 2014. 6884382 (Proceedings of the International Symposium on Consumer Electronics, ISCE).

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

TY - GEN

T1 - DPM

T2 - 18th IEEE International Symposium on Consumer Electronics, ISCE 2014

AU - Jo, Myung Hyun

AU - Ro, Won Woo

PY - 2014

Y1 - 2014

N2 - The MapReduce frameworks using a modern graphic processor (GPU) have improved the performance of data-intensive applications. While the prior researches have enhanced the parallelism of the MapReduce application on a GPU, archiving optimal distribution of big data on heterogeneous devices is still a challengeable issue. We therefore propose a method to evenly separate the computing cost under limited memory size. To solve this problem, we design and propose DPM, a Data Partitioning Method, using a GPU to smartly distribute workload of MapReduce. The proposed technique provides well-balanced processing cost for heterogeneous devices.

AB - The MapReduce frameworks using a modern graphic processor (GPU) have improved the performance of data-intensive applications. While the prior researches have enhanced the parallelism of the MapReduce application on a GPU, archiving optimal distribution of big data on heterogeneous devices is still a challengeable issue. We therefore propose a method to evenly separate the computing cost under limited memory size. To solve this problem, we design and propose DPM, a Data Partitioning Method, using a GPU to smartly distribute workload of MapReduce. The proposed technique provides well-balanced processing cost for heterogeneous devices.

UR - http://www.scopus.com/inward/record.url?scp=84907362941&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84907362941&partnerID=8YFLogxK

U2 - 10.1109/ISCE.2014.6884382

DO - 10.1109/ISCE.2014.6884382

M3 - Conference contribution

AN - SCOPUS:84907362941

SN - 9781479945924

T3 - Proceedings of the International Symposium on Consumer Electronics, ISCE

BT - ISCE 2014 - 18th IEEE International Symposium on Consumer Electronics

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 22 June 2014 through 25 June 2014

ER -