Analysis of the ParConnect algorithm ran on Intel Xeon Phi Knights Landing

Maximilian Hornung; Svilen Stefanov; David Schneller; Sharru Møller

doi:10.1016/j.parco.2017.08.011

Analysis of the ParConnect algorithm ran on Intel Xeon Phi Knights Landing

Maximilian Hornung
, Svilen Stefanov
, David Schneller
, Sharru Møller

Informatics 5 - Associate Professorship of Hardware-aware algorithms and software for HPC

Technical University of Munich

Research output: Contribution to journal › Article › peer-review

Abstract

Huge volume of data and high data complexity are the main challenges in metagenomic assembly. Therefore, Patrick Flick et al. present a novel highly-scalable distributed-memory parallel algorithm (ParConnect) to find connected subgraphs. Their approach to this subject leads to the well-studied problem of finding weakly connected components in a de Bruijn graph, which is used to represent overlaps between sequences of symbols. At the Student Cluster Competition 2016 (SCC 2016), we reproduced the results of their paper on an Intel Xeon Phi Knights Landing cluster and analyzed the share of communication time compared to the whole computation.

Original language	English
Pages (from-to)	46-53
Number of pages	8
Journal	Parallel Computing
Volume	70
DOIs	https://doi.org/10.1016/j.parco.2017.08.011
State	Published - Dec 2017

Keywords

Bioinformatics
De Bruijn Graph
Intel Xeon Phi
Reproducibility

Access to Document

10.1016/j.parco.2017.08.011

Cite this

@article{0aa3291841ef4b6e8dcc1d3b0550a4ab,

title = "Analysis of the ParConnect algorithm ran on Intel Xeon Phi Knights Landing",

abstract = "Huge volume of data and high data complexity are the main challenges in metagenomic assembly. Therefore, Patrick Flick et al. present a novel highly-scalable distributed-memory parallel algorithm (ParConnect) to find connected subgraphs. Their approach to this subject leads to the well-studied problem of finding weakly connected components in a de Bruijn graph, which is used to represent overlaps between sequences of symbols. At the Student Cluster Competition 2016 (SCC 2016), we reproduced the results of their paper on an Intel Xeon Phi Knights Landing cluster and analyzed the share of communication time compared to the whole computation.",

keywords = "Bioinformatics, De Bruijn Graph, Intel Xeon Phi, Reproducibility",

author = "Maximilian Hornung and Svilen Stefanov and David Schneller and Sharru M{\o}ller",

note = "Publisher Copyright: {\textcopyright} 2017 Elsevier B.V.",

year = "2017",

month = dec,

doi = "10.1016/j.parco.2017.08.011",

language = "English",

volume = "70",

pages = "46--53",

journal = "Parallel Computing",

issn = "0167-8191",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Analysis of the ParConnect algorithm ran on Intel Xeon Phi Knights Landing

AU - Hornung, Maximilian

AU - Stefanov, Svilen

AU - Schneller, David

AU - Møller, Sharru

PY - 2017/12

Y1 - 2017/12

N2 - Huge volume of data and high data complexity are the main challenges in metagenomic assembly. Therefore, Patrick Flick et al. present a novel highly-scalable distributed-memory parallel algorithm (ParConnect) to find connected subgraphs. Their approach to this subject leads to the well-studied problem of finding weakly connected components in a de Bruijn graph, which is used to represent overlaps between sequences of symbols. At the Student Cluster Competition 2016 (SCC 2016), we reproduced the results of their paper on an Intel Xeon Phi Knights Landing cluster and analyzed the share of communication time compared to the whole computation.

AB - Huge volume of data and high data complexity are the main challenges in metagenomic assembly. Therefore, Patrick Flick et al. present a novel highly-scalable distributed-memory parallel algorithm (ParConnect) to find connected subgraphs. Their approach to this subject leads to the well-studied problem of finding weakly connected components in a de Bruijn graph, which is used to represent overlaps between sequences of symbols. At the Student Cluster Competition 2016 (SCC 2016), we reproduced the results of their paper on an Intel Xeon Phi Knights Landing cluster and analyzed the share of communication time compared to the whole computation.

KW - Bioinformatics

KW - De Bruijn Graph

KW - Intel Xeon Phi

KW - Reproducibility

UR - https://www.scopus.com/pages/publications/85028886439

U2 - 10.1016/j.parco.2017.08.011

DO - 10.1016/j.parco.2017.08.011

M3 - Article

AN - SCOPUS:85028886439

SN - 0167-8191

VL - 70

SP - 46

EP - 53

JO - Parallel Computing

JF - Parallel Computing

ER -

Analysis of the ParConnect algorithm ran on Intel Xeon Phi Knights Landing

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this