Visualizing large 3D geodesic grid data with massively distributed GPUs

Jinrong Xie, Hongfeng Yu, Kwan Liu Maz

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Geodesic grids become increasingly prevalent in large weather and climate applications. The deluge amount of simulation data demands efficient and scalable visualization capabilities for scientific exploration and understanding. Given the unique characteristics of geodesic grids, no current techniques can scalably visualize scalar fields defined on a geodesic grid. In this paper, we present a new parallel ray-casting algorithm for large geodesic grids using massively distributed GPUs. We construct a spherical quadtree to adaptively partition and distribute the data according to the grid resolution of simulation, and ensure a balanced workload assignment over a large number of processors from different view angles. We have designed and implemented the entire rendering pipeline based on the MPI and CUDA architecture, and demonstrated the effectiveness and scalability of our approach using an example of large application on a supercomputer with thousands of GPUs.

Original languageEnglish (US)
Title of host publicationIEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings
EditorsHank Childs, Hank Childs, Renato Pajarola, Venkatram Vishwanath
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3-10
Number of pages8
ISBN (Electronic)9781479952151
DOIs
StatePublished - Jan 16 2014
Event4th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2014 - Paris, France
Duration: Oct 9 2014Oct 10 2014

Publication series

NameIEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings

Other

Other4th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2014
CountryFrance
CityParis
Period10/9/1410/10/14

Fingerprint

Supercomputers
Scalability
Casting
Visualization
Pipelines
Graphics processing unit

ASJC Scopus subject areas

  • Information Systems
  • Computer Vision and Pattern Recognition

Cite this

Xie, J., Yu, H., & Maz, K. L. (2014). Visualizing large 3D geodesic grid data with massively distributed GPUs. In H. Childs, H. Childs, R. Pajarola, & V. Vishwanath (Eds.), IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings (pp. 3-10). [7013198] (IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/LDAV.2014.7013198

Visualizing large 3D geodesic grid data with massively distributed GPUs. / Xie, Jinrong; Yu, Hongfeng; Maz, Kwan Liu.

IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings. ed. / Hank Childs; Hank Childs; Renato Pajarola; Venkatram Vishwanath. Institute of Electrical and Electronics Engineers Inc., 2014. p. 3-10 7013198 (IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Xie, J, Yu, H & Maz, KL 2014, Visualizing large 3D geodesic grid data with massively distributed GPUs. in H Childs, H Childs, R Pajarola & V Vishwanath (eds), IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings., 7013198, IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 3-10, 4th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2014, Paris, France, 10/9/14. https://doi.org/10.1109/LDAV.2014.7013198
Xie J, Yu H, Maz KL. Visualizing large 3D geodesic grid data with massively distributed GPUs. In Childs H, Childs H, Pajarola R, Vishwanath V, editors, IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2014. p. 3-10. 7013198. (IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings). https://doi.org/10.1109/LDAV.2014.7013198
Xie, Jinrong ; Yu, Hongfeng ; Maz, Kwan Liu. / Visualizing large 3D geodesic grid data with massively distributed GPUs. IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings. editor / Hank Childs ; Hank Childs ; Renato Pajarola ; Venkatram Vishwanath. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 3-10 (IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings).
@inproceedings{0f80ceadff89493ba5fb3fe240ef124d,
title = "Visualizing large 3D geodesic grid data with massively distributed GPUs",
abstract = "Geodesic grids become increasingly prevalent in large weather and climate applications. The deluge amount of simulation data demands efficient and scalable visualization capabilities for scientific exploration and understanding. Given the unique characteristics of geodesic grids, no current techniques can scalably visualize scalar fields defined on a geodesic grid. In this paper, we present a new parallel ray-casting algorithm for large geodesic grids using massively distributed GPUs. We construct a spherical quadtree to adaptively partition and distribute the data according to the grid resolution of simulation, and ensure a balanced workload assignment over a large number of processors from different view angles. We have designed and implemented the entire rendering pipeline based on the MPI and CUDA architecture, and demonstrated the effectiveness and scalability of our approach using an example of large application on a supercomputer with thousands of GPUs.",
author = "Jinrong Xie and Hongfeng Yu and Maz, {Kwan Liu}",
year = "2014",
month = "1",
day = "16",
doi = "10.1109/LDAV.2014.7013198",
language = "English (US)",
series = "IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "3--10",
editor = "Hank Childs and Hank Childs and Renato Pajarola and Venkatram Vishwanath",
booktitle = "IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings",

}

TY - GEN

T1 - Visualizing large 3D geodesic grid data with massively distributed GPUs

AU - Xie, Jinrong

AU - Yu, Hongfeng

AU - Maz, Kwan Liu

PY - 2014/1/16

Y1 - 2014/1/16

N2 - Geodesic grids become increasingly prevalent in large weather and climate applications. The deluge amount of simulation data demands efficient and scalable visualization capabilities for scientific exploration and understanding. Given the unique characteristics of geodesic grids, no current techniques can scalably visualize scalar fields defined on a geodesic grid. In this paper, we present a new parallel ray-casting algorithm for large geodesic grids using massively distributed GPUs. We construct a spherical quadtree to adaptively partition and distribute the data according to the grid resolution of simulation, and ensure a balanced workload assignment over a large number of processors from different view angles. We have designed and implemented the entire rendering pipeline based on the MPI and CUDA architecture, and demonstrated the effectiveness and scalability of our approach using an example of large application on a supercomputer with thousands of GPUs.

AB - Geodesic grids become increasingly prevalent in large weather and climate applications. The deluge amount of simulation data demands efficient and scalable visualization capabilities for scientific exploration and understanding. Given the unique characteristics of geodesic grids, no current techniques can scalably visualize scalar fields defined on a geodesic grid. In this paper, we present a new parallel ray-casting algorithm for large geodesic grids using massively distributed GPUs. We construct a spherical quadtree to adaptively partition and distribute the data according to the grid resolution of simulation, and ensure a balanced workload assignment over a large number of processors from different view angles. We have designed and implemented the entire rendering pipeline based on the MPI and CUDA architecture, and demonstrated the effectiveness and scalability of our approach using an example of large application on a supercomputer with thousands of GPUs.

UR - http://www.scopus.com/inward/record.url?scp=84946685182&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84946685182&partnerID=8YFLogxK

U2 - 10.1109/LDAV.2014.7013198

DO - 10.1109/LDAV.2014.7013198

M3 - Conference contribution

AN - SCOPUS:84946685182

T3 - IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings

SP - 3

EP - 10

BT - IEEE Symposium on Large Data Analysis and Visualization 2014, LDAV 2014 - Proceedings

A2 - Childs, Hank

A2 - Childs, Hank

A2 - Pajarola, Renato

A2 - Vishwanath, Venkatram

PB - Institute of Electrical and Electronics Engineers Inc.

ER -