• Sorted by Date • Classified by Publication Type • Classified by Research Category •
Stefan Zellmann, Ingo Wald, Joao Barbosa, Serkan Demirci, Alper Sahistan, and Uğur Güdükbay. Hybrid Image-/Data-Parallel Rendering Using Island Parallelism. In Proceedings of the IEEE 12th Symposium on Large Data Analysis and Visualization, pp. 1–10, LDAV '22, IEEE Computer Society, Washington, DC, USA, 6 December 2022.
In parallel ray tracing, techniques fall into one of two camps: image-parallel techniques aim at increasing frame rate by replicating scene data across nodes and splitting the rendering work across different ranks, and data-parallel techniques aim at increasing the size of the model that can be rendered by splitting the model across multiple ranks, but typically cannot scale much in frame rate. We propose and evaluate a hybrid approach that combines the advantages of both by splitting a set of N x M ranks into M islands of N ranks each and using data-parallel rendering within each island and image parallelism across islands. We discuss the integration of this concept into four wildly different parallel renderers and evaluate the efficacy of this approach based on multiple different data sets.
@InProceedings{ZellmannEtAl22, title = {{Hybrid Image-/Data-Parallel Rendering Using Island Parallelism}}, author = {Stefan Zellmann and Ingo Wald and Joao Barbosa and Serkan Demirci and Alper Sahistan and U{\^g}ur G{\"u}d{\"u}kbay}, booktitle = {Proceedings of the IEEE 12th Symposium on Large Data Analysis and Visualization}, pages = {1--10}, year = {2022}, month = {6 December}, editor = {Johanna Beyer and Steffen Frey and Paul Navr{\' a}til}, series = {LDAV '22}, publisher = {IEEE Computer Society}, address = {Washington, DC, USA}, bib2html_dl_pdf = "http://www.cs.bilkent.edu.tr/~gudukbay/publications/papers/conf_papers/Zellmann_Et_Al_LDAV_2022.pdf", bib2html_pubtype = {Refereed Conference Papers}, bib2html_rescat = {Computer Graphics}, pdf = {https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9966396&tag=1}, url = {https://ieeexplore.ieee.org/document/9966396}, ee = {https://doi.org/10.1109/LDAV57265.2022.9966396}, abstract = {In parallel ray tracing, techniques fall into one of two camps: image-parallel techniques aim at increasing frame rate by replicating scene data across nodes and splitting the rendering work across different ranks, and data-parallel techniques aim at increasing the size of the model that can be rendered by splitting the model across multiple ranks, but typically cannot scale much in frame rate. We propose and evaluate a hybrid approach that combines the advantages of both by splitting a set of N x M ranks into M islands of N ranks each and using data-parallel rendering within each island and image parallelism across islands. We discuss the integration of this concept into four wildly different parallel renderers and evaluate the efficacy of this approach based on multiple different data sets.} }
Generated by bib2html.pl (written by Patrick Riley ) on Sun Apr 21, 2024 11:32:41