Efficient parallel optimization of volume meshes on heterogeneous computing systems

Zuofu Cheng, Eric Gene Shaffer, Raine Yeh, George Zagaris, Luke Olson

Research output: Contribution to journalArticle

Abstract

We describe a parallel algorithmic framework for optimizing the shape of elements in a simplicial volume mesh. Using fine-grained parallelism and asymmetric multiprocessing on multi-core CPU and modern graphics processing unit hardware simultaneously, we achieve speedups of more than tenfold over current state-of-the-art serial methods. In addition, improved mesh quality is obtained by optimizing both the surface and the interior vertex positions in a single pass, using feature preservation to maintain fidelity to the original mesh geometry. The framework is flexible in terms of the core numerical optimization method employed, and we provide performance results for both gradient-based and derivative-free optimization methods.

Original languageEnglish (US)
Pages (from-to)717-726
Number of pages10
JournalEngineering with Computers
Volume33
Issue number4
DOIs
StatePublished - Oct 1 2017

Fingerprint

Parallel Optimization
Heterogeneous Computing
Optimization Methods
Derivative-free Optimization
Mesh
Derivative-free Methods
Multiprocessing
Mesh Quality
Numerical Optimization
Graphics Processing Unit
Preservation
Fidelity
Computer hardware
Parallelism
Program processors
Interior
Numerical Methods
Hardware
Gradient
Derivatives

Keywords

  • GPU applications
  • Mesh optimization
  • Parallel algorithms

ASJC Scopus subject areas

  • Software
  • Modeling and Simulation
  • Engineering(all)
  • Computer Science Applications

Cite this

Efficient parallel optimization of volume meshes on heterogeneous computing systems. / Cheng, Zuofu; Shaffer, Eric Gene; Yeh, Raine; Zagaris, George; Olson, Luke.

In: Engineering with Computers, Vol. 33, No. 4, 01.10.2017, p. 717-726.

Research output: Contribution to journalArticle

@article{fbab19f15be440949f7c678d36ce7780,
title = "Efficient parallel optimization of volume meshes on heterogeneous computing systems",
abstract = "We describe a parallel algorithmic framework for optimizing the shape of elements in a simplicial volume mesh. Using fine-grained parallelism and asymmetric multiprocessing on multi-core CPU and modern graphics processing unit hardware simultaneously, we achieve speedups of more than tenfold over current state-of-the-art serial methods. In addition, improved mesh quality is obtained by optimizing both the surface and the interior vertex positions in a single pass, using feature preservation to maintain fidelity to the original mesh geometry. The framework is flexible in terms of the core numerical optimization method employed, and we provide performance results for both gradient-based and derivative-free optimization methods.",
keywords = "GPU applications, Mesh optimization, Parallel algorithms",
author = "Zuofu Cheng and Shaffer, {Eric Gene} and Raine Yeh and George Zagaris and Luke Olson",
year = "2017",
month = "10",
day = "1",
doi = "10.1007/s00366-014-0393-7",
language = "English (US)",
volume = "33",
pages = "717--726",
journal = "Engineering with Computers",
issn = "0177-0667",
publisher = "Springer London",
number = "4",

}

TY - JOUR

T1 - Efficient parallel optimization of volume meshes on heterogeneous computing systems

AU - Cheng, Zuofu

AU - Shaffer, Eric Gene

AU - Yeh, Raine

AU - Zagaris, George

AU - Olson, Luke

PY - 2017/10/1

Y1 - 2017/10/1

N2 - We describe a parallel algorithmic framework for optimizing the shape of elements in a simplicial volume mesh. Using fine-grained parallelism and asymmetric multiprocessing on multi-core CPU and modern graphics processing unit hardware simultaneously, we achieve speedups of more than tenfold over current state-of-the-art serial methods. In addition, improved mesh quality is obtained by optimizing both the surface and the interior vertex positions in a single pass, using feature preservation to maintain fidelity to the original mesh geometry. The framework is flexible in terms of the core numerical optimization method employed, and we provide performance results for both gradient-based and derivative-free optimization methods.

AB - We describe a parallel algorithmic framework for optimizing the shape of elements in a simplicial volume mesh. Using fine-grained parallelism and asymmetric multiprocessing on multi-core CPU and modern graphics processing unit hardware simultaneously, we achieve speedups of more than tenfold over current state-of-the-art serial methods. In addition, improved mesh quality is obtained by optimizing both the surface and the interior vertex positions in a single pass, using feature preservation to maintain fidelity to the original mesh geometry. The framework is flexible in terms of the core numerical optimization method employed, and we provide performance results for both gradient-based and derivative-free optimization methods.

KW - GPU applications

KW - Mesh optimization

KW - Parallel algorithms

UR - http://www.scopus.com/inward/record.url?scp=85029646552&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029646552&partnerID=8YFLogxK

U2 - 10.1007/s00366-014-0393-7

DO - 10.1007/s00366-014-0393-7

M3 - Article

AN - SCOPUS:85029646552

VL - 33

SP - 717

EP - 726

JO - Engineering with Computers

JF - Engineering with Computers

SN - 0177-0667

IS - 4

ER -