Hybrid Online Autotuning for Parallel Ray Tracing

Simple item page
dc.contributor.authorHerveau, Killianen_US
dc.contributor.authorPfaffe, Philipen_US
dc.contributor.authorTillmann, Martin Peteren_US
dc.contributor.authorTichy, Walter F.en_US
dc.contributor.authorDachsbacher, Carstenen_US
dc.contributor.editorChilds, Hank and Frey, Steffenen_US
dc.date.accessioned2019-06-02T18:25:59Z
dc.date.available2019-06-02T18:25:59Z
dc.date.issued2019
dc.description.abstractAcceleration structures are key to high performance parallel ray tracing. Maximizing performance requires configuring the degrees of freedom (e.g., construction parameters) these data structures expose. Whether a parameter setting is optimal depends on the input (e.g., the scene and view parameters) and hardware. Manual selection is tedious, error prone, and is not portable. To automate the parameter selection task we use a hybrid of model-based prediction and online autotuning. The combination benefits from the best of both worlds: one-shot configuration selection when inputs are known or similar, effective exploration of the configuration space otherwise. Online tuning additionally serves to train the model on real inputs without requiring a-priori training samples. Online autotuning outperforms best-practice configurations recommended by the literature, by up to 11% median. The model predictions achieve 95% of the online autotuning performance while reducing 90% of the autotuner overhead. Hybrid online autotuning thus enables always-on tuning of parallel ray tracing.en_US
dc.description.sectionheadersSession 2
dc.description.seriesinformationEurographics Symposium on Parallel Graphics and Visualization
dc.identifier.doi10.2312/pgv.20191110
dc.identifier.isbn978-3-03868-079-6
dc.identifier.issn1727-348X
dc.identifier.pages59-68
dc.identifier.urihttps://doi.org/10.2312/pgv.20191110
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/pgv20191110
dc.publisherThe Eurographics Associationen_US
dc.subjectI.3.6 [Computer Graphics]
dc.subjectMethodology and Techniques
dc.subjectGraphics data structures and data types
dc.titleHybrid Online Autotuning for Parallel Ray Tracingen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
059-068.pdf
Size:
372.35 KB
Format:
Adobe Portable Document Format
Download
Collections
EGPGV19: Eurographics Symposium on Parallel Graphics and Visualization