Visualizing large-scale parallel communication traces using a particle animation technique

Carmen Sigovan, Chris W. Muelder, Kwan-Liu Ma

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Large-scale scientific simulations require execution on parallel computing systems in order to yield useful results in a reasonable time frame. But parallel execution adds communication overhead. The impact that this overhead has on performance may be difficult to gauge, as parallel application behaviors are typically harder to understand than the sequential types. We introduce an animation-based interactive visualization technique for the analysis of communication patterns occurring in parallel application execution. Our method has the advantages of illustrating the dynamic communication patterns in the system as well as a static image of MPI (Message Passing Interface) utilization history. We also devise a data streaming mechanism that allows for the exploration of very large data sets. We demonstrate the effectiveness of our approach scaling up to 16 thousand processes using a series of trace data sets of ScaLAPACK matrix operations functions.

Original languageEnglish (US)
Pages (from-to)141-150
Number of pages10
JournalComputer Graphics Forum
Volume32
Issue number3 PART2
DOIs
StatePublished - Jan 1 2013

Fingerprint

Animation
Communication
Message passing
Parallel processing systems
Gages
Visualization

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Graphics and Computer-Aided Design

Cite this

Visualizing large-scale parallel communication traces using a particle animation technique. / Sigovan, Carmen; Muelder, Chris W.; Ma, Kwan-Liu.

In: Computer Graphics Forum, Vol. 32, No. 3 PART2, 01.01.2013, p. 141-150.

Research output: Contribution to journalArticle

Sigovan, Carmen ; Muelder, Chris W. ; Ma, Kwan-Liu. / Visualizing large-scale parallel communication traces using a particle animation technique. In: Computer Graphics Forum. 2013 ; Vol. 32, No. 3 PART2. pp. 141-150.
@article{ae5c73c304d142c6a5f332e47fbf4c03,
title = "Visualizing large-scale parallel communication traces using a particle animation technique",
abstract = "Large-scale scientific simulations require execution on parallel computing systems in order to yield useful results in a reasonable time frame. But parallel execution adds communication overhead. The impact that this overhead has on performance may be difficult to gauge, as parallel application behaviors are typically harder to understand than the sequential types. We introduce an animation-based interactive visualization technique for the analysis of communication patterns occurring in parallel application execution. Our method has the advantages of illustrating the dynamic communication patterns in the system as well as a static image of MPI (Message Passing Interface) utilization history. We also devise a data streaming mechanism that allows for the exploration of very large data sets. We demonstrate the effectiveness of our approach scaling up to 16 thousand processes using a series of trace data sets of ScaLAPACK matrix operations functions.",
author = "Carmen Sigovan and Muelder, {Chris W.} and Kwan-Liu Ma",
year = "2013",
month = "1",
day = "1",
doi = "10.1111/cgf.12101",
language = "English (US)",
volume = "32",
pages = "141--150",
journal = "Computer Graphics Forum",
issn = "0167-7055",
publisher = "Wiley-Blackwell",
number = "3 PART2",

}

TY - JOUR

T1 - Visualizing large-scale parallel communication traces using a particle animation technique

AU - Sigovan, Carmen

AU - Muelder, Chris W.

AU - Ma, Kwan-Liu

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Large-scale scientific simulations require execution on parallel computing systems in order to yield useful results in a reasonable time frame. But parallel execution adds communication overhead. The impact that this overhead has on performance may be difficult to gauge, as parallel application behaviors are typically harder to understand than the sequential types. We introduce an animation-based interactive visualization technique for the analysis of communication patterns occurring in parallel application execution. Our method has the advantages of illustrating the dynamic communication patterns in the system as well as a static image of MPI (Message Passing Interface) utilization history. We also devise a data streaming mechanism that allows for the exploration of very large data sets. We demonstrate the effectiveness of our approach scaling up to 16 thousand processes using a series of trace data sets of ScaLAPACK matrix operations functions.

AB - Large-scale scientific simulations require execution on parallel computing systems in order to yield useful results in a reasonable time frame. But parallel execution adds communication overhead. The impact that this overhead has on performance may be difficult to gauge, as parallel application behaviors are typically harder to understand than the sequential types. We introduce an animation-based interactive visualization technique for the analysis of communication patterns occurring in parallel application execution. Our method has the advantages of illustrating the dynamic communication patterns in the system as well as a static image of MPI (Message Passing Interface) utilization history. We also devise a data streaming mechanism that allows for the exploration of very large data sets. We demonstrate the effectiveness of our approach scaling up to 16 thousand processes using a series of trace data sets of ScaLAPACK matrix operations functions.

UR - http://www.scopus.com/inward/record.url?scp=84879779337&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879779337&partnerID=8YFLogxK

U2 - 10.1111/cgf.12101

DO - 10.1111/cgf.12101

M3 - Article

AN - SCOPUS:84879779337

VL - 32

SP - 141

EP - 150

JO - Computer Graphics Forum

JF - Computer Graphics Forum

SN - 0167-7055

IS - 3 PART2

ER -