Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures

Research output: Contribution to conferencePaper

45 Citations (Scopus)

Abstract

As computing technology continues to advance, computational modeling of scientific and engineering problems produces data of increasing complexity: large in size and unstructural in shape. Volume visualization of such data is a challenging problem. This paper proposes a distributed parallel solution that makes ray-casting volume rendering of unstructured-grid data practical. Both the data and the rendering process are distributed among processors. At each processor, ray-casting of local data is performed independent of the other processors. The global image compositing processes, which require inter-processor communication, are overlapped with the local ray-casting processes to achieve maximum parallel efficiency. This algorithm differs from previous ones in four ways: it is completely distributed, less view-dependent, reasonably scalable, and flexible. Without dynamic load balancing, test results on the Intel Paragon using from two to 128 processors show, on average, about 60% parallel efficiency.

Original languageEnglish (US)
Pages23-30
Number of pages8
StatePublished - Dec 1 1995
Externally publishedYes
EventProceedings of the 1995 Parallel Rendering Symposium - Atlanta, GA, USA
Duration: Oct 30 1995Oct 31 1995

Other

OtherProceedings of the 1995 Parallel Rendering Symposium
CityAtlanta, GA, USA
Period10/30/9510/31/95

Fingerprint

Memory architecture
Casting
Volume rendering
Dynamic loads
Resource allocation
Visualization
Communication

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Ma, K-L. (1995). Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures. 23-30. Paper presented at Proceedings of the 1995 Parallel Rendering Symposium, Atlanta, GA, USA, .

Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures. / Ma, Kwan-Liu.

1995. 23-30 Paper presented at Proceedings of the 1995 Parallel Rendering Symposium, Atlanta, GA, USA, .

Research output: Contribution to conferencePaper

Ma, K-L 1995, 'Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures', Paper presented at Proceedings of the 1995 Parallel Rendering Symposium, Atlanta, GA, USA, 10/30/95 - 10/31/95 pp. 23-30.
Ma K-L. Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures. 1995. Paper presented at Proceedings of the 1995 Parallel Rendering Symposium, Atlanta, GA, USA, .
Ma, Kwan-Liu. / Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures. Paper presented at Proceedings of the 1995 Parallel Rendering Symposium, Atlanta, GA, USA, .8 p.
@conference{903a2cbb6d5f464ba63562fc59c85a91,
title = "Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures",
abstract = "As computing technology continues to advance, computational modeling of scientific and engineering problems produces data of increasing complexity: large in size and unstructural in shape. Volume visualization of such data is a challenging problem. This paper proposes a distributed parallel solution that makes ray-casting volume rendering of unstructured-grid data practical. Both the data and the rendering process are distributed among processors. At each processor, ray-casting of local data is performed independent of the other processors. The global image compositing processes, which require inter-processor communication, are overlapped with the local ray-casting processes to achieve maximum parallel efficiency. This algorithm differs from previous ones in four ways: it is completely distributed, less view-dependent, reasonably scalable, and flexible. Without dynamic load balancing, test results on the Intel Paragon using from two to 128 processors show, on average, about 60{\%} parallel efficiency.",
author = "Kwan-Liu Ma",
year = "1995",
month = "12",
day = "1",
language = "English (US)",
pages = "23--30",
note = "Proceedings of the 1995 Parallel Rendering Symposium ; Conference date: 30-10-1995 Through 31-10-1995",

}

TY - CONF

T1 - Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures

AU - Ma, Kwan-Liu

PY - 1995/12/1

Y1 - 1995/12/1

N2 - As computing technology continues to advance, computational modeling of scientific and engineering problems produces data of increasing complexity: large in size and unstructural in shape. Volume visualization of such data is a challenging problem. This paper proposes a distributed parallel solution that makes ray-casting volume rendering of unstructured-grid data practical. Both the data and the rendering process are distributed among processors. At each processor, ray-casting of local data is performed independent of the other processors. The global image compositing processes, which require inter-processor communication, are overlapped with the local ray-casting processes to achieve maximum parallel efficiency. This algorithm differs from previous ones in four ways: it is completely distributed, less view-dependent, reasonably scalable, and flexible. Without dynamic load balancing, test results on the Intel Paragon using from two to 128 processors show, on average, about 60% parallel efficiency.

AB - As computing technology continues to advance, computational modeling of scientific and engineering problems produces data of increasing complexity: large in size and unstructural in shape. Volume visualization of such data is a challenging problem. This paper proposes a distributed parallel solution that makes ray-casting volume rendering of unstructured-grid data practical. Both the data and the rendering process are distributed among processors. At each processor, ray-casting of local data is performed independent of the other processors. The global image compositing processes, which require inter-processor communication, are overlapped with the local ray-casting processes to achieve maximum parallel efficiency. This algorithm differs from previous ones in four ways: it is completely distributed, less view-dependent, reasonably scalable, and flexible. Without dynamic load balancing, test results on the Intel Paragon using from two to 128 processors show, on average, about 60% parallel efficiency.

UR - http://www.scopus.com/inward/record.url?scp=0029544349&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029544349&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0029544349

SP - 23

EP - 30

ER -