In situ visualization at extreme scale: Challenges and opportunities

Research output: Contribution to journalArticle

96 Citations (Scopus)

Abstract

Scientific computing at the petascale level enables us to answer many difficult scientific questions, but the resulting data are too large to store and study directly with conventional postprocessing visualization tools. This problem will only become more severe as we reach exascale computing. A plausible, attractive solution involves processing data in situ with the simulation to reduce the data that must be transferred over networks and stored and to prepare the data for more cost-effective postprocessing visualization. The data could be reduced with compression, feature extraction, and visualization methods. This article discusses critical issues in realizing in situ visualization and data reduction and suggests important research directions.

Original languageEnglish (US)
Article number5307638
Pages (from-to)14-19
Number of pages6
JournalIEEE Computer Graphics and Applications
Volume29
Issue number6
DOIs
StatePublished - Nov 1 2009

Fingerprint

Visualization
Natural sciences computing
Feature extraction
Data reduction
Costs

Keywords

  • Computer graphics
  • Scalability
  • Scientific discovery
  • Supercomputing
  • Visualization

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

In situ visualization at extreme scale : Challenges and opportunities. / Ma, Kwan-Liu.

In: IEEE Computer Graphics and Applications, Vol. 29, No. 6, 5307638, 01.11.2009, p. 14-19.

Research output: Contribution to journalArticle

@article{3bc0f774cc69440db2261922eb6ab38e,
title = "In situ visualization at extreme scale: Challenges and opportunities",
abstract = "Scientific computing at the petascale level enables us to answer many difficult scientific questions, but the resulting data are too large to store and study directly with conventional postprocessing visualization tools. This problem will only become more severe as we reach exascale computing. A plausible, attractive solution involves processing data in situ with the simulation to reduce the data that must be transferred over networks and stored and to prepare the data for more cost-effective postprocessing visualization. The data could be reduced with compression, feature extraction, and visualization methods. This article discusses critical issues in realizing in situ visualization and data reduction and suggests important research directions.",
keywords = "Computer graphics, Scalability, Scientific discovery, Supercomputing, Visualization",
author = "Kwan-Liu Ma",
year = "2009",
month = "11",
day = "1",
doi = "10.1109/MCG.2009.120",
language = "English (US)",
volume = "29",
pages = "14--19",
journal = "IEEE Computer Graphics and Applications",
issn = "0272-1716",
publisher = "IEEE Computer Society",
number = "6",

}

TY - JOUR

T1 - In situ visualization at extreme scale

T2 - Challenges and opportunities

AU - Ma, Kwan-Liu

PY - 2009/11/1

Y1 - 2009/11/1

N2 - Scientific computing at the petascale level enables us to answer many difficult scientific questions, but the resulting data are too large to store and study directly with conventional postprocessing visualization tools. This problem will only become more severe as we reach exascale computing. A plausible, attractive solution involves processing data in situ with the simulation to reduce the data that must be transferred over networks and stored and to prepare the data for more cost-effective postprocessing visualization. The data could be reduced with compression, feature extraction, and visualization methods. This article discusses critical issues in realizing in situ visualization and data reduction and suggests important research directions.

AB - Scientific computing at the petascale level enables us to answer many difficult scientific questions, but the resulting data are too large to store and study directly with conventional postprocessing visualization tools. This problem will only become more severe as we reach exascale computing. A plausible, attractive solution involves processing data in situ with the simulation to reduce the data that must be transferred over networks and stored and to prepare the data for more cost-effective postprocessing visualization. The data could be reduced with compression, feature extraction, and visualization methods. This article discusses critical issues in realizing in situ visualization and data reduction and suggests important research directions.

KW - Computer graphics

KW - Scalability

KW - Scientific discovery

KW - Supercomputing

KW - Visualization

UR - http://www.scopus.com/inward/record.url?scp=70449514481&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449514481&partnerID=8YFLogxK

U2 - 10.1109/MCG.2009.120

DO - 10.1109/MCG.2009.120

M3 - Article

AN - SCOPUS:70449514481

VL - 29

SP - 14

EP - 19

JO - IEEE Computer Graphics and Applications

JF - IEEE Computer Graphics and Applications

SN - 0272-1716

IS - 6

M1 - 5307638

ER -