Optimal pipeline decomposition and adaptive network mapping to support distributed remote visualization

Michelle Zhu, Qishi Wu, Nageswara S.V. Rao, Sitharama Iyengar

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

This paper discusses algorithmic and implementation aspects of a distributed remote visualization system that optimally decomposes and adaptively maps the visualization pipeline to a wide-area shared or dedicated network. The first node of the system typically generates or stores raw data sets, and a remote client resides on the last node equipped with a display device ranging from a personal desktop to a powerwall. Intermediate nodes include workstations, clusters, or rendering engines, which can be located anywhere on the network. We employ a regression method to estimate the effective bandwidth of a transport path. Based on link measurements, node characteristics, and module properties, we strategically organize visualization pipeline modules into groups and dynamically assign the groups to various network nodes to achieve minimal total delay or maximal frame rate. We propose polynomial-time algorithms using the dynamic programming method to compute optimal solutions for the problems of pipeline decomposition and network mapping under different constraint conditions. The proposed remote visualization system is implemented and deployed at several geographically distributed nodes for experimental testing. The proposed decomposition and mapping scheme is generic and can be applied to other distributed applications whose computing components form a linear arrangement.

Original languageEnglish
Pages (from-to)947-956
Number of pages10
JournalJournal of Parallel and Distributed Computing
Volume67
Issue number8
DOIs
StatePublished - 1 Aug 2007

Fingerprint

Visualization
Pipelines
Decomposition
Decompose
Vertex of a graph
Dynamic programming
Effective Bandwidth
Telecommunication links
Module
Display devices
Distributed Applications
Polynomials
Engines
Bandwidth
Rendering
Polynomial-time Algorithm
Dynamic Programming
Assign
Arrangement
Display

Keywords

  • Bandwidth measurement
  • Distributed computing
  • Dynamic programming
  • Network mapping
  • Remote visualization
  • Visualization pipeline

Cite this

@article{cc93f7a217894fcfa4b4a3fb969293f5,
title = "Optimal pipeline decomposition and adaptive network mapping to support distributed remote visualization",
abstract = "This paper discusses algorithmic and implementation aspects of a distributed remote visualization system that optimally decomposes and adaptively maps the visualization pipeline to a wide-area shared or dedicated network. The first node of the system typically generates or stores raw data sets, and a remote client resides on the last node equipped with a display device ranging from a personal desktop to a powerwall. Intermediate nodes include workstations, clusters, or rendering engines, which can be located anywhere on the network. We employ a regression method to estimate the effective bandwidth of a transport path. Based on link measurements, node characteristics, and module properties, we strategically organize visualization pipeline modules into groups and dynamically assign the groups to various network nodes to achieve minimal total delay or maximal frame rate. We propose polynomial-time algorithms using the dynamic programming method to compute optimal solutions for the problems of pipeline decomposition and network mapping under different constraint conditions. The proposed remote visualization system is implemented and deployed at several geographically distributed nodes for experimental testing. The proposed decomposition and mapping scheme is generic and can be applied to other distributed applications whose computing components form a linear arrangement.",
keywords = "Bandwidth measurement, Distributed computing, Dynamic programming, Network mapping, Remote visualization, Visualization pipeline",
author = "Michelle Zhu and Qishi Wu and Rao, {Nageswara S.V.} and Sitharama Iyengar",
year = "2007",
month = "8",
day = "1",
doi = "10.1016/j.jpdc.2007.04.007",
language = "English",
volume = "67",
pages = "947--956",
journal = "Journal of Parallel and Distributed Computing",
issn = "0743-7315",
publisher = "Academic Press Inc.",
number = "8",

}

Optimal pipeline decomposition and adaptive network mapping to support distributed remote visualization. / Zhu, Michelle; Wu, Qishi; Rao, Nageswara S.V.; Iyengar, Sitharama.

In: Journal of Parallel and Distributed Computing, Vol. 67, No. 8, 01.08.2007, p. 947-956.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Optimal pipeline decomposition and adaptive network mapping to support distributed remote visualization

AU - Zhu, Michelle

AU - Wu, Qishi

AU - Rao, Nageswara S.V.

AU - Iyengar, Sitharama

PY - 2007/8/1

Y1 - 2007/8/1

N2 - This paper discusses algorithmic and implementation aspects of a distributed remote visualization system that optimally decomposes and adaptively maps the visualization pipeline to a wide-area shared or dedicated network. The first node of the system typically generates or stores raw data sets, and a remote client resides on the last node equipped with a display device ranging from a personal desktop to a powerwall. Intermediate nodes include workstations, clusters, or rendering engines, which can be located anywhere on the network. We employ a regression method to estimate the effective bandwidth of a transport path. Based on link measurements, node characteristics, and module properties, we strategically organize visualization pipeline modules into groups and dynamically assign the groups to various network nodes to achieve minimal total delay or maximal frame rate. We propose polynomial-time algorithms using the dynamic programming method to compute optimal solutions for the problems of pipeline decomposition and network mapping under different constraint conditions. The proposed remote visualization system is implemented and deployed at several geographically distributed nodes for experimental testing. The proposed decomposition and mapping scheme is generic and can be applied to other distributed applications whose computing components form a linear arrangement.

AB - This paper discusses algorithmic and implementation aspects of a distributed remote visualization system that optimally decomposes and adaptively maps the visualization pipeline to a wide-area shared or dedicated network. The first node of the system typically generates or stores raw data sets, and a remote client resides on the last node equipped with a display device ranging from a personal desktop to a powerwall. Intermediate nodes include workstations, clusters, or rendering engines, which can be located anywhere on the network. We employ a regression method to estimate the effective bandwidth of a transport path. Based on link measurements, node characteristics, and module properties, we strategically organize visualization pipeline modules into groups and dynamically assign the groups to various network nodes to achieve minimal total delay or maximal frame rate. We propose polynomial-time algorithms using the dynamic programming method to compute optimal solutions for the problems of pipeline decomposition and network mapping under different constraint conditions. The proposed remote visualization system is implemented and deployed at several geographically distributed nodes for experimental testing. The proposed decomposition and mapping scheme is generic and can be applied to other distributed applications whose computing components form a linear arrangement.

KW - Bandwidth measurement

KW - Distributed computing

KW - Dynamic programming

KW - Network mapping

KW - Remote visualization

KW - Visualization pipeline

UR - http://www.scopus.com/inward/record.url?scp=34447285514&partnerID=8YFLogxK

U2 - 10.1016/j.jpdc.2007.04.007

DO - 10.1016/j.jpdc.2007.04.007

M3 - Article

AN - SCOPUS:34447285514

VL - 67

SP - 947

EP - 956

JO - Journal of Parallel and Distributed Computing

JF - Journal of Parallel and Distributed Computing

SN - 0743-7315

IS - 8

ER -