Parallel generation of architecture on the GPU

Markus Steinberger, Michael Kenzel, Bernhard Kainz, Jörg Müller, Wonka Peter, Dieter Schmalstieg

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

In this paper, we present a novel approach for the parallel evaluation of procedural shape grammars on the graphics processing unit (GPU). Unlike previous approaches that are either limited in the kind of shapes they allow, the amount of parallelism they can take advantage of, or both, our method supports state of the art procedural modeling including stochasticity and context-sensitivity. To increase parallelism, we explicitly express independence in the grammar, reduce inter-rule dependencies required for context-sensitive evaluation, and introduce intra-rule parallelism. Our rule scheduling scheme avoids unnecessary back and forth between CPU and GPU and reduces round trips to slow global memory by dynamically grouping rules in on-chip shared memory. Our GPU shape grammar implementation is multiple orders of magnitude faster than the standard in CPU-based rule evaluation, while offering equal expressive power. In comparison to the state of the art in GPU shape grammar derivation, our approach is nearly 50 times faster, while adding support for geometric context-sensitivity.

Original languageEnglish (US)
Pages (from-to)73-82
Number of pages10
JournalComputer Graphics Forum
Volume33
Issue number2
DOIs
StatePublished - 2014
Externally publishedYes

Fingerprint

Program processors
Data storage equipment
Scheduling
Graphics processing unit

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Steinberger, M., Kenzel, M., Kainz, B., Müller, J., Peter, W., & Schmalstieg, D. (2014). Parallel generation of architecture on the GPU. Computer Graphics Forum, 33(2), 73-82. https://doi.org/10.1111/cgf.12312

Parallel generation of architecture on the GPU. / Steinberger, Markus; Kenzel, Michael; Kainz, Bernhard; Müller, Jörg; Peter, Wonka; Schmalstieg, Dieter.

In: Computer Graphics Forum, Vol. 33, No. 2, 2014, p. 73-82.

Research output: Contribution to journalArticle

Steinberger, M, Kenzel, M, Kainz, B, Müller, J, Peter, W & Schmalstieg, D 2014, 'Parallel generation of architecture on the GPU', Computer Graphics Forum, vol. 33, no. 2, pp. 73-82. https://doi.org/10.1111/cgf.12312
Steinberger M, Kenzel M, Kainz B, Müller J, Peter W, Schmalstieg D. Parallel generation of architecture on the GPU. Computer Graphics Forum. 2014;33(2):73-82. https://doi.org/10.1111/cgf.12312
Steinberger, Markus ; Kenzel, Michael ; Kainz, Bernhard ; Müller, Jörg ; Peter, Wonka ; Schmalstieg, Dieter. / Parallel generation of architecture on the GPU. In: Computer Graphics Forum. 2014 ; Vol. 33, No. 2. pp. 73-82.
@article{4f825fdbc63049358e611a98ec58c259,
title = "Parallel generation of architecture on the GPU",
abstract = "In this paper, we present a novel approach for the parallel evaluation of procedural shape grammars on the graphics processing unit (GPU). Unlike previous approaches that are either limited in the kind of shapes they allow, the amount of parallelism they can take advantage of, or both, our method supports state of the art procedural modeling including stochasticity and context-sensitivity. To increase parallelism, we explicitly express independence in the grammar, reduce inter-rule dependencies required for context-sensitive evaluation, and introduce intra-rule parallelism. Our rule scheduling scheme avoids unnecessary back and forth between CPU and GPU and reduces round trips to slow global memory by dynamically grouping rules in on-chip shared memory. Our GPU shape grammar implementation is multiple orders of magnitude faster than the standard in CPU-based rule evaluation, while offering equal expressive power. In comparison to the state of the art in GPU shape grammar derivation, our approach is nearly 50 times faster, while adding support for geometric context-sensitivity.",
author = "Markus Steinberger and Michael Kenzel and Bernhard Kainz and J{\"o}rg M{\"u}ller and Wonka Peter and Dieter Schmalstieg",
year = "2014",
doi = "10.1111/cgf.12312",
language = "English (US)",
volume = "33",
pages = "73--82",
journal = "Computer Graphics Forum",
issn = "0167-7055",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - Parallel generation of architecture on the GPU

AU - Steinberger, Markus

AU - Kenzel, Michael

AU - Kainz, Bernhard

AU - Müller, Jörg

AU - Peter, Wonka

AU - Schmalstieg, Dieter

PY - 2014

Y1 - 2014

N2 - In this paper, we present a novel approach for the parallel evaluation of procedural shape grammars on the graphics processing unit (GPU). Unlike previous approaches that are either limited in the kind of shapes they allow, the amount of parallelism they can take advantage of, or both, our method supports state of the art procedural modeling including stochasticity and context-sensitivity. To increase parallelism, we explicitly express independence in the grammar, reduce inter-rule dependencies required for context-sensitive evaluation, and introduce intra-rule parallelism. Our rule scheduling scheme avoids unnecessary back and forth between CPU and GPU and reduces round trips to slow global memory by dynamically grouping rules in on-chip shared memory. Our GPU shape grammar implementation is multiple orders of magnitude faster than the standard in CPU-based rule evaluation, while offering equal expressive power. In comparison to the state of the art in GPU shape grammar derivation, our approach is nearly 50 times faster, while adding support for geometric context-sensitivity.

AB - In this paper, we present a novel approach for the parallel evaluation of procedural shape grammars on the graphics processing unit (GPU). Unlike previous approaches that are either limited in the kind of shapes they allow, the amount of parallelism they can take advantage of, or both, our method supports state of the art procedural modeling including stochasticity and context-sensitivity. To increase parallelism, we explicitly express independence in the grammar, reduce inter-rule dependencies required for context-sensitive evaluation, and introduce intra-rule parallelism. Our rule scheduling scheme avoids unnecessary back and forth between CPU and GPU and reduces round trips to slow global memory by dynamically grouping rules in on-chip shared memory. Our GPU shape grammar implementation is multiple orders of magnitude faster than the standard in CPU-based rule evaluation, while offering equal expressive power. In comparison to the state of the art in GPU shape grammar derivation, our approach is nearly 50 times faster, while adding support for geometric context-sensitivity.

UR - http://www.scopus.com/inward/record.url?scp=84901852471&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901852471&partnerID=8YFLogxK

U2 - 10.1111/cgf.12312

DO - 10.1111/cgf.12312

M3 - Article

AN - SCOPUS:84901852471

VL - 33

SP - 73

EP - 82

JO - Computer Graphics Forum

JF - Computer Graphics Forum

SN - 0167-7055

IS - 2

ER -