Accelerating 3D Digital Differential Analyzer Ray Tracing Algorithm on the GPU Using CUDA

We present a novel optimized 3DDDA ray tracing algorithm on GPUs with CUDA architecture. The grid data is built on the host and then copied to the device. The whole traversing procedure is done on the GPU device. So it just needs the data transformation between the host and the device once. The results show that the CUDA implementation parallel code using different block size on GPU achieves 2.9x to 5.9x faster speed compared with the CPU implementation. And the performance is significantly influenced by the size of the CUDA block. 3DDDA algorithm on CUDA architecture can really get the performance accelerated with a reasonable block size, but it is difficult to play high-end GPU performance. The main reason is that 3DDDA algorithm for memory access is discontinuous, so it is difficult to play high-frequency performance of RAM DDR5.

[1]  Aaron Knoll,et al.  Ray tracing animated scenes using coherent grid traversal , 2006, SIGGRAPH 2006.

[2]  Thierry Priol,et al.  An Efficient Parallel Ray Tracing Scheme for Highly Parallel Architectures , 1990, Advances in Computer Graphics Hardware V.

[3]  Pat Hanrahan,et al.  Ray tracing on programmable graphics hardware , 2002, SIGGRAPH Courses.

[4]  David P. Luebke,et al.  CUDA: Scalable parallel programming for high-performance scientific computing , 2008, 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[5]  Thomas Wiegand,et al.  SIFT Implementation and Optimization for General-Purpose GPU , 2007 .

[6]  Philipp Slusallek,et al.  Efficient stack-less BVH traversal for ray tracing , 2011, SCC.

[7]  Martín Pedemonte,et al.  Improving the Performance of a Ray Tracing Algorithm Using a GPU , 2010, 2010 XXIX International Conference of the Chilean Computer Science Society.

[8]  Philipp Slusallek,et al.  A parallel algorithm for construction of uniform grids , 2009, High Performance Graphics.

[9]  Ying Tan,et al.  GPU-based parallel particle swarm optimization , 2009, 2009 IEEE Congress on Evolutionary Computation.

[10]  Turner Whitted,et al.  An improved illumination model for shaded display , 1979, SIGGRAPH Courses.

[11]  Arno Formella,et al.  Ray-tracing acceleration techniques , 2000 .

[12]  Takayuki Tanaka,et al.  ARTS: Accelerated Ray-Tracing System , 1986, IEEE Computer Graphics and Applications.

[13]  Bo Zhou,et al.  GPU-accelerated Monte Carlo convolution/superposition implementation for dose calculation. , 2010, Medical physics.

[14]  Turner Whitted,et al.  An improved illumination model for shaded display , 1979, SIGGRAPH.

[15]  Marc Levoy,et al.  Fitting smooth surfaces to dense polygon meshes , 1996, SIGGRAPH.

[16]  Vlastimil Havran,et al.  Ray Tracing on a GPU with CUDA – Comparative Study of Three Algorithms , 2010 .

[17]  Mary K. Vernon,et al.  Toward a multicore architecture for real-time ray-tracing , 2008, 2008 41st IEEE/ACM International Symposium on Microarchitecture.

[18]  P. J. Narayanan,et al.  Accelerating Large Graph Algorithms on the GPU Using CUDA , 2007, HiPC.