Visualization and parallel I/O at extreme scale

R. B. Ross, T. Peterka, H. W. Shen, Y. Hong, Kwan-Liu Ma, H. Yu, K. Moreland

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

In our efforts to solve ever more challenging problems through computational techniques, the scale of our compute systems continues to grow. As we approach petascale, it becomes increasingly important that all the resources in the system be used as efficiently as possible, not just the floating-point units. Because of hardware, software, and usability challenges, storage resources are often one of the most poorly used and performing components of today's compute systems. This situation can be especially true in the case of the analysis phases of scientific workflows. In this paper we discuss the impact of large-scale data on visual analysis operations and examine a collection of approaches to I/O in the visual analysis process. First we examine the performance of volume rendering on a leadership-computing platform and assess the relative cost of I/O, rendering, and compositing operations. Next we analyze the performance implications of eliminating preprocessing from this example workflow. Then we describe a technique that uses data reorganization to improve access times for data-intensive volume rendering.

Original languageEnglish (US)
Article number012099
JournalJournal of Physics: Conference Series
Volume125
DOIs
StatePublished - Jan 1 2008

Fingerprint

resources
access time
leadership
preprocessing
floating
hardware
platforms
costs
computer programs

ASJC Scopus subject areas

  • Physics and Astronomy(all)

Cite this

Ross, R. B., Peterka, T., Shen, H. W., Hong, Y., Ma, K-L., Yu, H., & Moreland, K. (2008). Visualization and parallel I/O at extreme scale. Journal of Physics: Conference Series, 125, [012099]. https://doi.org/10.1088/1742-6596/125/1/012099

Visualization and parallel I/O at extreme scale. / Ross, R. B.; Peterka, T.; Shen, H. W.; Hong, Y.; Ma, Kwan-Liu; Yu, H.; Moreland, K.

In: Journal of Physics: Conference Series, Vol. 125, 012099, 01.01.2008.

Research output: Contribution to journalArticle

Ross, R. B. ; Peterka, T. ; Shen, H. W. ; Hong, Y. ; Ma, Kwan-Liu ; Yu, H. ; Moreland, K. / Visualization and parallel I/O at extreme scale. In: Journal of Physics: Conference Series. 2008 ; Vol. 125.
@article{355d0b62a3dc4855b8bd255e2f72b00c,
title = "Visualization and parallel I/O at extreme scale",
abstract = "In our efforts to solve ever more challenging problems through computational techniques, the scale of our compute systems continues to grow. As we approach petascale, it becomes increasingly important that all the resources in the system be used as efficiently as possible, not just the floating-point units. Because of hardware, software, and usability challenges, storage resources are often one of the most poorly used and performing components of today's compute systems. This situation can be especially true in the case of the analysis phases of scientific workflows. In this paper we discuss the impact of large-scale data on visual analysis operations and examine a collection of approaches to I/O in the visual analysis process. First we examine the performance of volume rendering on a leadership-computing platform and assess the relative cost of I/O, rendering, and compositing operations. Next we analyze the performance implications of eliminating preprocessing from this example workflow. Then we describe a technique that uses data reorganization to improve access times for data-intensive volume rendering.",
author = "Ross, {R. B.} and T. Peterka and Shen, {H. W.} and Y. Hong and Kwan-Liu Ma and H. Yu and K. Moreland",
year = "2008",
month = "1",
day = "1",
doi = "10.1088/1742-6596/125/1/012099",
language = "English (US)",
volume = "125",
journal = "Journal of Physics: Conference Series",
issn = "1742-6588",
publisher = "IOP Publishing Ltd.",

}

TY - JOUR

T1 - Visualization and parallel I/O at extreme scale

AU - Ross, R. B.

AU - Peterka, T.

AU - Shen, H. W.

AU - Hong, Y.

AU - Ma, Kwan-Liu

AU - Yu, H.

AU - Moreland, K.

PY - 2008/1/1

Y1 - 2008/1/1

N2 - In our efforts to solve ever more challenging problems through computational techniques, the scale of our compute systems continues to grow. As we approach petascale, it becomes increasingly important that all the resources in the system be used as efficiently as possible, not just the floating-point units. Because of hardware, software, and usability challenges, storage resources are often one of the most poorly used and performing components of today's compute systems. This situation can be especially true in the case of the analysis phases of scientific workflows. In this paper we discuss the impact of large-scale data on visual analysis operations and examine a collection of approaches to I/O in the visual analysis process. First we examine the performance of volume rendering on a leadership-computing platform and assess the relative cost of I/O, rendering, and compositing operations. Next we analyze the performance implications of eliminating preprocessing from this example workflow. Then we describe a technique that uses data reorganization to improve access times for data-intensive volume rendering.

AB - In our efforts to solve ever more challenging problems through computational techniques, the scale of our compute systems continues to grow. As we approach petascale, it becomes increasingly important that all the resources in the system be used as efficiently as possible, not just the floating-point units. Because of hardware, software, and usability challenges, storage resources are often one of the most poorly used and performing components of today's compute systems. This situation can be especially true in the case of the analysis phases of scientific workflows. In this paper we discuss the impact of large-scale data on visual analysis operations and examine a collection of approaches to I/O in the visual analysis process. First we examine the performance of volume rendering on a leadership-computing platform and assess the relative cost of I/O, rendering, and compositing operations. Next we analyze the performance implications of eliminating preprocessing from this example workflow. Then we describe a technique that uses data reorganization to improve access times for data-intensive volume rendering.

UR - http://www.scopus.com/inward/record.url?scp=65549159997&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=65549159997&partnerID=8YFLogxK

U2 - 10.1088/1742-6596/125/1/012099

DO - 10.1088/1742-6596/125/1/012099

M3 - Article

VL - 125

JO - Journal of Physics: Conference Series

JF - Journal of Physics: Conference Series

SN - 1742-6588

M1 - 012099

ER -