Building of SBVH on Graphical Hardware

This thesis describes the design and implementation of the SBVH build algorithm using the parallel CUDA architecture on the GPU. It also compares this implementation with a serial implementation of the same build algorithm as well as with other acceleration structures. The measurements performed across eight 3D scenes reveal that the CUDA variant of the SBVH build is substantially faster (on average 213×) than the serial variant while maintaining the same acceleration structure quality. Compared with other GPU acceleration structures, the GPU SBVH offers a better acceleration performance (on average 41% to 57%) for the cost of a longer build process (on average 4.0× to 4.4×).

[1]  Maurice Herlihy,et al.  The Art of Multiprocessor Programming, Revised Reprint , 2012 .

[2]  Vlastimil Havran,et al.  Review: Kd‐tree Traversal Algorithms for Ray Tracing , 2011, Comput. Graph. Forum.

[3]  I. Wald,et al.  On building fast kd-Trees for Ray Tracing, and on doing that in O(N log N) , 2006, 2006 IEEE Symposium on Interactive Ray Tracing.

[4]  Andreas Dietrich,et al.  Spatial splits in bounding volume hierarchies , 2009, High Performance Graphics.

[5]  M. Steinberger,et al.  ScatterAlloc: Massively parallel dynamic memory allocation for the GPU , 2012, 2012 Innovative Parallel Computing (InPar).

[6]  Greg Humphreys,et al.  Physically Based Rendering, Second Edition: From Theory To Implementation , 2010 .

[7]  Donald P. Greenberg,et al.  Improved Computational Methods for Ray Tracing , 1984, TOGS.

[8]  Vlastimil Havran,et al.  Register Efficient Dynamic Memory Allocator for GPUs , 2015, Comput. Graph. Forum.

[9]  Vlastimil Havran,et al.  Massively Parallel Hierarchical Scene Processing with Applications in Rendering , 2013, Comput. Graph. Forum.

[10]  G. Greiner,et al.  Early Split Clipping for Bounding Volume Hierarchies , 2007, 2007 IEEE Symposium on Interactive Ray Tracing.

[11]  Michael Goesele,et al.  Fast dynamic memory allocator for massively parallel architectures , 2013, GPGPU@ASPLOS.

[12]  Timo Aila,et al.  Understanding the efficiency of ray traversal on GPUs , 2009, High Performance Graphics.

[13]  I. Wald,et al.  On fast Construction of SAH-based Bounding Volume Hierarchies , 2007, 2007 IEEE Symposium on Interactive Ray Tracing.

[14]  Sven Woop,et al.  A Ray Tracing Hardware Architecture for Dynamic Scenes , 2004 .

[15]  John Salmon,et al.  Automatic Creation of Object Hierarchies for Ray Tracing , 1987, IEEE Computer Graphics and Applications.

[16]  A. Keller,et al.  The edge volume heuristic - robust triangle subdivision for improved BVH performance , 2008, 2008 IEEE Symposium on Interactive Ray Tracing.