Skip to Main content Skip to Navigation
Journal articles

Parallel extraction and simplification of large isosurfaces using an extended tandem algorithm

Abstract : In order to deal with the common trend in size increase of volumetric datasets, in the past few years research in isosurface extraction has focused on related aspects such as surface simplification and load-balanced parallel algorithms. We present a parallel, block-wise extension of the tandem algorithm by Attali et al., which simplifies on the fly an isosurface being extracted. Our approach minimizes the overall memory consumption using an adequate block splitting and merging strategy along with the introduction of a component dumping mechanism that drastically reduces the amount of memory needed for particular datasets such as those encountered in geophysics. As soon as detected, surface components are migrated to the disk along with a meta-data index (oriented bounding box, volume, etc.) that permits further improved exploration scenarios (small component removal or particularly oriented component selection for instance). For ease of implementation, we carefully describe a master and worker algorithm architecture that clearly separates the four required basic tasks. We show several results of our parallel algorithm applied on a geophysical dataset of size 7000 × 1600 × 2000.
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download

https://hal.inria.fr/inria-00436423
Contributor : Dimitri Komatitsch <>
Submitted on : Friday, November 12, 2010 - 6:24:49 PM
Last modification on : Tuesday, April 21, 2020 - 2:00:42 AM
Document(s) archivé(s) le : Sunday, February 13, 2011 - 2:25:25 AM

File

2010_dupuy_jobard_guillon_kesk...
Files produced by the author(s)

Identifiers

Collections

Citation

Guilhem Dupuy, Bruno Jobard, Sebastien Guillon, Noomane Keskes, Dimitri Komatitsch. Parallel extraction and simplification of large isosurfaces using an extended tandem algorithm. Computer-Aided Design, Elsevier, 2010, 42 (2), pp.129-138. ⟨10.1016/j.cad.2009.04.016⟩. ⟨inria-00436423⟩

Share

Metrics

Record views

683

Files downloads

934