Multi-kernel ray traversal for graphics processing units

Thomas Schiffer, Dieter W. Fellner

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

Ray tracing is a very popular family of algorithms that are used to compute images with high visual quality. One of its core challenges is designing an efficient mapping of ray traversal computations to massively parallel hardware architectures like modern algorithms graphics processing units (GPUs). In this paper we investigate the performance of state-of-the-art ray traversal algorithms on GPUs and discuss their potentials and limitations. Based on this analysis, a novel ray traversal scheme called batch tracing is proposed. It subdivides the task into multiple kernels, each of which is designed for efficient parallel execution. Our algorithm achieves comparable performance to current approaches and represents a promising direction for future research.

Original languageEnglish
Title of host publicationCommunications in Computer and Information Science
PublisherSpringer Verlag
Pages78-93
Number of pages16
Volume550
ISBN (Print)9783319251165
DOIs
Publication statusPublished - 2015
EventInternational Conference on Computer Vision Theory and Applications - Lissabon, Portugal
Duration: 5 Jan 20148 Jan 2014

Publication series

NameCommunications in Computer and Information Science
Volume550
ISSN (Print)18650929

Conference

ConferenceInternational Conference on Computer Vision Theory and Applications
CountryPortugal
CityLissabon
Period5/01/148/01/14

    Fingerprint

Keywords

  • Graphics processing units
  • Parallelism
  • Ray tracing
  • SIMT

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Schiffer, T., & Fellner, D. W. (2015). Multi-kernel ray traversal for graphics processing units. In Communications in Computer and Information Science (Vol. 550, pp. 78-93). (Communications in Computer and Information Science; Vol. 550). Springer Verlag. https://doi.org/10.1007/978-3-319-25117-2_6