Efficient parallel volume rendering of large-scale adaptive mesh refinement data

Nick Leaf, Venkatram Vishwanath, Joseph Insley, Mark Hereld, Michael E. Papka, Kwan-Liu Ma

Research output: Contribution to conferencePaper

6 Citations (Scopus)

Abstract

Adaptive Mesh Refinement is a popular approach for allocating scarce computing resources to the most important portions of the simulation domain. This approach implies spatial compression and the large simulation sizes which necessitate it. We present a novel, cluster- and GPU-parallel rendering scheme for AMR data, which is built on previous work in the GPU ray casting of AMR data. Our approach utilizes the existing AMR structure to subdivide the problem into convexly-bounded chunks and perform static load-balancing. We take advantage of data locality within chunks to interpolate directly between blocks without the need to store ghost cells on the interior boundaries. We also present a novel block decomposition method, and analyze its performance against two alternative methods. Finally, we examine the interactivity of our renderer for multiple datasets, and consider its scalability across a large number of GPUs.

Original languageEnglish (US)
Pages35-42
Number of pages8
DOIs
StatePublished - Jan 1 2013
Event2013 3rd IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2013 - Atlanta, GA, United States
Duration: Oct 13 2013Oct 14 2013

Other

Other2013 3rd IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2013
CountryUnited States
CityAtlanta, GA
Period10/13/1310/14/13

Fingerprint

Volume rendering
Resource allocation
Scalability
Casting
Decomposition
Graphics processing unit

Keywords

  • adaptive mesh refinement
  • Volume rendering

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Leaf, N., Vishwanath, V., Insley, J., Hereld, M., Papka, M. E., & Ma, K-L. (2013). Efficient parallel volume rendering of large-scale adaptive mesh refinement data. 35-42. Paper presented at 2013 3rd IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2013, Atlanta, GA, United States. https://doi.org/10.1109/LDAV.2013.6675156

Efficient parallel volume rendering of large-scale adaptive mesh refinement data. / Leaf, Nick; Vishwanath, Venkatram; Insley, Joseph; Hereld, Mark; Papka, Michael E.; Ma, Kwan-Liu.

2013. 35-42 Paper presented at 2013 3rd IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2013, Atlanta, GA, United States.

Research output: Contribution to conferencePaper

Leaf, N, Vishwanath, V, Insley, J, Hereld, M, Papka, ME & Ma, K-L 2013, 'Efficient parallel volume rendering of large-scale adaptive mesh refinement data' Paper presented at 2013 3rd IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2013, Atlanta, GA, United States, 10/13/13 - 10/14/13, pp. 35-42. https://doi.org/10.1109/LDAV.2013.6675156
Leaf N, Vishwanath V, Insley J, Hereld M, Papka ME, Ma K-L. Efficient parallel volume rendering of large-scale adaptive mesh refinement data. 2013. Paper presented at 2013 3rd IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2013, Atlanta, GA, United States. https://doi.org/10.1109/LDAV.2013.6675156
Leaf, Nick ; Vishwanath, Venkatram ; Insley, Joseph ; Hereld, Mark ; Papka, Michael E. ; Ma, Kwan-Liu. / Efficient parallel volume rendering of large-scale adaptive mesh refinement data. Paper presented at 2013 3rd IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2013, Atlanta, GA, United States.8 p.
@conference{eafe0b0c80b14da286ddd6f28ae8135f,
title = "Efficient parallel volume rendering of large-scale adaptive mesh refinement data",
abstract = "Adaptive Mesh Refinement is a popular approach for allocating scarce computing resources to the most important portions of the simulation domain. This approach implies spatial compression and the large simulation sizes which necessitate it. We present a novel, cluster- and GPU-parallel rendering scheme for AMR data, which is built on previous work in the GPU ray casting of AMR data. Our approach utilizes the existing AMR structure to subdivide the problem into convexly-bounded chunks and perform static load-balancing. We take advantage of data locality within chunks to interpolate directly between blocks without the need to store ghost cells on the interior boundaries. We also present a novel block decomposition method, and analyze its performance against two alternative methods. Finally, we examine the interactivity of our renderer for multiple datasets, and consider its scalability across a large number of GPUs.",
keywords = "adaptive mesh refinement, Volume rendering",
author = "Nick Leaf and Venkatram Vishwanath and Joseph Insley and Mark Hereld and Papka, {Michael E.} and Kwan-Liu Ma",
year = "2013",
month = "1",
day = "1",
doi = "10.1109/LDAV.2013.6675156",
language = "English (US)",
pages = "35--42",
note = "2013 3rd IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2013 ; Conference date: 13-10-2013 Through 14-10-2013",

}

TY - CONF

T1 - Efficient parallel volume rendering of large-scale adaptive mesh refinement data

AU - Leaf, Nick

AU - Vishwanath, Venkatram

AU - Insley, Joseph

AU - Hereld, Mark

AU - Papka, Michael E.

AU - Ma, Kwan-Liu

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Adaptive Mesh Refinement is a popular approach for allocating scarce computing resources to the most important portions of the simulation domain. This approach implies spatial compression and the large simulation sizes which necessitate it. We present a novel, cluster- and GPU-parallel rendering scheme for AMR data, which is built on previous work in the GPU ray casting of AMR data. Our approach utilizes the existing AMR structure to subdivide the problem into convexly-bounded chunks and perform static load-balancing. We take advantage of data locality within chunks to interpolate directly between blocks without the need to store ghost cells on the interior boundaries. We also present a novel block decomposition method, and analyze its performance against two alternative methods. Finally, we examine the interactivity of our renderer for multiple datasets, and consider its scalability across a large number of GPUs.

AB - Adaptive Mesh Refinement is a popular approach for allocating scarce computing resources to the most important portions of the simulation domain. This approach implies spatial compression and the large simulation sizes which necessitate it. We present a novel, cluster- and GPU-parallel rendering scheme for AMR data, which is built on previous work in the GPU ray casting of AMR data. Our approach utilizes the existing AMR structure to subdivide the problem into convexly-bounded chunks and perform static load-balancing. We take advantage of data locality within chunks to interpolate directly between blocks without the need to store ghost cells on the interior boundaries. We also present a novel block decomposition method, and analyze its performance against two alternative methods. Finally, we examine the interactivity of our renderer for multiple datasets, and consider its scalability across a large number of GPUs.

KW - adaptive mesh refinement

KW - Volume rendering

UR - http://www.scopus.com/inward/record.url?scp=84899089535&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899089535&partnerID=8YFLogxK

U2 - 10.1109/LDAV.2013.6675156

DO - 10.1109/LDAV.2013.6675156

M3 - Paper

SP - 35

EP - 42

ER -