A highly parallelized decoder for random network coding leveraging GPGPU

Joon Sang Park, Seung Jun Baek, Kyogu Lee

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Network coding has been shown to improve various performance metrics in computer networks. However, the use of network coding, especially random linear network coding, incurs serious time delay in the decoding process and thus it is imperative to use a network coding implementation that has low decoding latency characteristics, e.g. a parallelized implementation. In this paper, we investigate the problem of parallelizing Pipeline network coding, a variant of random linear coding recently developed in order to alleviate the problems of random linear coding. We propose a novel massively parallelized decoding algorithm leveraging General Purpose Graphics Processing Unit (GPGPU) and show its performance enhancement by up to 100% compared with previous GPGPU-based parallel algorithms via experiments on real systems.

Original languageEnglish
Pages (from-to)233-240
Number of pages8
JournalComputer Journal
Volume57
Issue number2
DOIs
Publication statusPublished - 2014 Feb 1

Fingerprint

Network coding
Decoding
Linear networks
Computer networks
Parallel algorithms
Time delay
Pipelines
Graphics processing unit
Experiments

Keywords

  • GPGPU
  • Network coding
  • Parallelization

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

A highly parallelized decoder for random network coding leveraging GPGPU. / Park, Joon Sang; Baek, Seung Jun; Lee, Kyogu.

In: Computer Journal, Vol. 57, No. 2, 01.02.2014, p. 233-240.

Research output: Contribution to journalArticle

Park, Joon Sang ; Baek, Seung Jun ; Lee, Kyogu. / A highly parallelized decoder for random network coding leveraging GPGPU. In: Computer Journal. 2014 ; Vol. 57, No. 2. pp. 233-240.
@article{2c0cdf4c5d53402d98faf1a41f44dbe4,
title = "A highly parallelized decoder for random network coding leveraging GPGPU",
abstract = "Network coding has been shown to improve various performance metrics in computer networks. However, the use of network coding, especially random linear network coding, incurs serious time delay in the decoding process and thus it is imperative to use a network coding implementation that has low decoding latency characteristics, e.g. a parallelized implementation. In this paper, we investigate the problem of parallelizing Pipeline network coding, a variant of random linear coding recently developed in order to alleviate the problems of random linear coding. We propose a novel massively parallelized decoding algorithm leveraging General Purpose Graphics Processing Unit (GPGPU) and show its performance enhancement by up to 100{\%} compared with previous GPGPU-based parallel algorithms via experiments on real systems.",
keywords = "GPGPU, Network coding, Parallelization",
author = "Park, {Joon Sang} and Baek, {Seung Jun} and Kyogu Lee",
year = "2014",
month = "2",
day = "1",
doi = "10.1093/comjnl/bxs173",
language = "English",
volume = "57",
pages = "233--240",
journal = "Computer Journal",
issn = "0010-4620",
publisher = "Oxford University Press",
number = "2",

}

TY - JOUR

T1 - A highly parallelized decoder for random network coding leveraging GPGPU

AU - Park, Joon Sang

AU - Baek, Seung Jun

AU - Lee, Kyogu

PY - 2014/2/1

Y1 - 2014/2/1

N2 - Network coding has been shown to improve various performance metrics in computer networks. However, the use of network coding, especially random linear network coding, incurs serious time delay in the decoding process and thus it is imperative to use a network coding implementation that has low decoding latency characteristics, e.g. a parallelized implementation. In this paper, we investigate the problem of parallelizing Pipeline network coding, a variant of random linear coding recently developed in order to alleviate the problems of random linear coding. We propose a novel massively parallelized decoding algorithm leveraging General Purpose Graphics Processing Unit (GPGPU) and show its performance enhancement by up to 100% compared with previous GPGPU-based parallel algorithms via experiments on real systems.

AB - Network coding has been shown to improve various performance metrics in computer networks. However, the use of network coding, especially random linear network coding, incurs serious time delay in the decoding process and thus it is imperative to use a network coding implementation that has low decoding latency characteristics, e.g. a parallelized implementation. In this paper, we investigate the problem of parallelizing Pipeline network coding, a variant of random linear coding recently developed in order to alleviate the problems of random linear coding. We propose a novel massively parallelized decoding algorithm leveraging General Purpose Graphics Processing Unit (GPGPU) and show its performance enhancement by up to 100% compared with previous GPGPU-based parallel algorithms via experiments on real systems.

KW - GPGPU

KW - Network coding

KW - Parallelization

UR - http://www.scopus.com/inward/record.url?scp=84893258887&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893258887&partnerID=8YFLogxK

U2 - 10.1093/comjnl/bxs173

DO - 10.1093/comjnl/bxs173

M3 - Article

VL - 57

SP - 233

EP - 240

JO - Computer Journal

JF - Computer Journal

SN - 0010-4620

IS - 2

ER -