TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

Fast Hardware Construction and Refitting of Quantized Bounding Volume Hierarchies

Tutkimustuotosvertaisarvioitu

Standard

Fast Hardware Construction and Refitting of Quantized Bounding Volume Hierarchies. / Viitanen, Timo; Koskela, Matias; Jääskeläinen, Pekka; Immonen, Kalle; Takala, Jarmo.

julkaisussa: Computer Graphics Forum, Vuosikerta 36, Nro 4, 05.07.2017, s. 167-178.

Tutkimustuotosvertaisarvioitu

Harvard

Viitanen, T, Koskela, M, Jääskeläinen, P, Immonen, K & Takala, J 2017, 'Fast Hardware Construction and Refitting of Quantized Bounding Volume Hierarchies', Computer Graphics Forum, Vuosikerta. 36, Nro 4, Sivut 167-178. https://doi.org/10.1111/cgf.13233

APA

Vancouver

Author

Viitanen, Timo ; Koskela, Matias ; Jääskeläinen, Pekka ; Immonen, Kalle ; Takala, Jarmo. / Fast Hardware Construction and Refitting of Quantized Bounding Volume Hierarchies. Julkaisussa: Computer Graphics Forum. 2017 ; Vuosikerta 36, Nro 4. Sivut 167-178.

Bibtex - Lataa

@article{838335358d9a4bdc85ce84b47507c0bf,
title = "Fast Hardware Construction and Refitting of Quantized Bounding Volume Hierarchies",
abstract = "There is recent interest in GPU architectures designed to accelerate ray tracing, especially on mobile systems with limited memory bandwidth. A promising recent approach is to store and traverse Bounding Volume Hierarchies (BVHs), used to accelerate ray tracing, in low arithmetic precision. However, so far there is no research on refitting or construction of such compressed BVHs, which is necessary for any scenes with dynamic content. We find that in a hardware-accelerated tree update, significant memory traffic and runtime savings are available from streaming, bottom-up compression. Novel algorithmic techniques of modulo encoding and treelet-based compression are proposed to reduce backtracking inherent in bottom-up compression. Together, these techniques reduce backtracking to a small fraction. Compared to a separate top-down compression pass, streaming bottom-up compression with the proposed optimizations saves on average 42{\%} of memory accesses for LBVH construction and 56{\%} for refitting of compressed BVHs, over 16 test scenes. In architectural simulation, the proposed streaming compression reduces LBVH runtime by 20{\%} compared to a single-precision build, and 41{\%} compared to a single-precision build followed by top-down compression. Since memory traffic dominates the energy cost of refitting and LBVH construction, energy consumption is expected to fall by a similar fraction.",
author = "Timo Viitanen and Matias Koskela and Pekka J{\"a}{\"a}skel{\"a}inen and Kalle Immonen and Jarmo Takala",
year = "2017",
month = "7",
day = "5",
doi = "10.1111/cgf.13233",
language = "English",
volume = "36",
pages = "167--178",
journal = "Computer Graphics Forum",
issn = "0167-7055",
publisher = "Wiley",
number = "4",

}

RIS (suitable for import to EndNote) - Lataa

TY - JOUR

T1 - Fast Hardware Construction and Refitting of Quantized Bounding Volume Hierarchies

AU - Viitanen, Timo

AU - Koskela, Matias

AU - Jääskeläinen, Pekka

AU - Immonen, Kalle

AU - Takala, Jarmo

PY - 2017/7/5

Y1 - 2017/7/5

N2 - There is recent interest in GPU architectures designed to accelerate ray tracing, especially on mobile systems with limited memory bandwidth. A promising recent approach is to store and traverse Bounding Volume Hierarchies (BVHs), used to accelerate ray tracing, in low arithmetic precision. However, so far there is no research on refitting or construction of such compressed BVHs, which is necessary for any scenes with dynamic content. We find that in a hardware-accelerated tree update, significant memory traffic and runtime savings are available from streaming, bottom-up compression. Novel algorithmic techniques of modulo encoding and treelet-based compression are proposed to reduce backtracking inherent in bottom-up compression. Together, these techniques reduce backtracking to a small fraction. Compared to a separate top-down compression pass, streaming bottom-up compression with the proposed optimizations saves on average 42% of memory accesses for LBVH construction and 56% for refitting of compressed BVHs, over 16 test scenes. In architectural simulation, the proposed streaming compression reduces LBVH runtime by 20% compared to a single-precision build, and 41% compared to a single-precision build followed by top-down compression. Since memory traffic dominates the energy cost of refitting and LBVH construction, energy consumption is expected to fall by a similar fraction.

AB - There is recent interest in GPU architectures designed to accelerate ray tracing, especially on mobile systems with limited memory bandwidth. A promising recent approach is to store and traverse Bounding Volume Hierarchies (BVHs), used to accelerate ray tracing, in low arithmetic precision. However, so far there is no research on refitting or construction of such compressed BVHs, which is necessary for any scenes with dynamic content. We find that in a hardware-accelerated tree update, significant memory traffic and runtime savings are available from streaming, bottom-up compression. Novel algorithmic techniques of modulo encoding and treelet-based compression are proposed to reduce backtracking inherent in bottom-up compression. Together, these techniques reduce backtracking to a small fraction. Compared to a separate top-down compression pass, streaming bottom-up compression with the proposed optimizations saves on average 42% of memory accesses for LBVH construction and 56% for refitting of compressed BVHs, over 16 test scenes. In architectural simulation, the proposed streaming compression reduces LBVH runtime by 20% compared to a single-precision build, and 41% compared to a single-precision build followed by top-down compression. Since memory traffic dominates the energy cost of refitting and LBVH construction, energy consumption is expected to fall by a similar fraction.

U2 - 10.1111/cgf.13233

DO - 10.1111/cgf.13233

M3 - Article

VL - 36

SP - 167

EP - 178

JO - Computer Graphics Forum

JF - Computer Graphics Forum

SN - 0167-7055

IS - 4

ER -