Abstract
A CUDA kernel will utilize high-latency local memory for storage when there are not enough registers to hold the required data or if the data is an array that is accessed using a variable index within a loop. However, accesses from local memory take longer than accesses from registers and shared memory, so it is desirable to minimize the use of local memory. This paper contains an analysis of strategies used to reduce the use of local memory in a CUDA implementation of belief propagation for stereo processing. We perform experiments using registers as well as shared memory as alternate locations for data initially placed in local memory, and then develop a hybrid implementation that allows the programmer to store an adjustable amount of data in shared, register, and local memory. We show results of running our optimized implementations on two different stereo sets and across three generations of nVidia GPUs, and introduce an auto-tuning implementation that generates an optimized belief propagation implementation on any input stereo set on any CUDA-capable GPU.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
NVIDIA CUDA Programming Guide: Version 3.0. NVIDIA Corporation (2010), http://developer.nvidia.com/cuda
Brunton, A., Shu, C., Roth, G.: Belief propagation on the gpu for stereo vision, June 2006, pp. 76–76 (2006)
Callahan, D., Carr, S., Kennedy, K.: Improving register allocation for subscripted variables. SIGPLAN Not. 25(6), 53–65 (1990)
Cooper, K.D., Lu, J.: Register promotion in c programs. In: Proc. ACM SIGPLAN Conf. Programming Language Design and Implementation (PLDI 1997), pp. 308–319. ACM Press, New York (1997)
Datta, K., Murphy, M., Volkov, V., Williams, S., Carter, J., Oliker, L., Patterson, D., Shalf, J., Yelick, K.: Stencil computation optimization and auto-tuning on state-of-the-art multicore architectures. In: SC 2008: Proceedings of the 2008 ACM/IEEE Conference on Supercomputing, pp. 1–12. IEEE Press, Piscataway (2008)
Felzenszwalb, P.F., Huttenlocher, D.P.: Efficient belief propagation for early vision. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 261–268 (2004)
Grauer-Gray, S., Kambhamettu, C.: Hierarchical belief propagation to reduce search space using cuda for stereo and motion estimation. In: 2009 IEEE Workshop on Applications of Computer Vision, WACV 2009 (2009)
Grauer-Gray, S., Kambhamettu, C., Palaniappan, K.: Gpu implementation of belief propagation using cuda for cloud tracking and reconstruction. pp. 1–4 (2008)
Li, Y., Dongarra, J., Tomov, S.: A note on auto-tuning GEMM for gPUs. In: Allen, G., Nabrzyski, J., Seidel, E., van Albada, G.D., Dongarra, J., Sloot, P.M.A. (eds.) ICCS 2009. LNCS, vol. 5544, pp. 884–892. Springer, Heidelberg (2009)
Liang, C.K., Cheng, C.C., Lai, Y.C., Chen, L.G., Chen, H.H.: Hardware-efficient belief propagation. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 80–87 (2009)
Liu, Y., Zhang, E.Z., Shen, X.: A cross-input adaptive framework for gpu program optimizations. In: International Parallel and Distributed Processing Symposium, pp. 1–10 (2009)
Nukada, A., Matsuoka, S.: Auto-tuning 3-d fft library for cuda gpus. In: SC 2009: Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, pp. 1–10. ACM, New York (2009)
Ryoo, S., Rodrigues, C.I., Baghsorkhi, S.S., Stone, S.S., Kirk, D.B., Hwu, W.m.W.: Optimization principles and application performance evaluation of a multithreaded gpu using cuda. In: PPoPP 2008: Proceedings of the 13th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, pp. 73–82. ACM Press, New York (2008)
Scharstein, D., Szeliski, R., Zabih, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms, pp. 131–140 (2001)
Scharstein, D., Szeliski, R.: High-accuracy stereo depth maps using structured light. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, p. 195 (2003)
Sun, J., Zheng, N.N., Shum, H.Y.: Stereo matching using belief propagation. IEEE Trans. Pattern Anal. Mach. Intell. 25(7), 787–800 (2003)
Ivanchenko, V., Shen, H., Coughlan, J.: Elevation-based stereo implemented in real-time on a gpu. In: 2009 IEEE Workshop on Applications of Computer Vision, WACV 2009 (2009)
Yang, Q., Wang, L., Yang, R., Wang, S., Liao, M., Nistér, D.: Real-time global stereo matching using hierarchical belief propagation. In: British Machine Vision Conf., pp. 989–998 (2006)
Xu, Y., Chen, H., Klette, R., Liu, J., Vaudrey, T.: Belief propagation implementation using cuda on an nvidia gtx 280, pp. 180–189 (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Grauer-Gray, S., Cavazos, J. (2011). Optimizing and Auto-tuning Belief Propagation on the GPU. In: Cooper, K., Mellor-Crummey, J., Sarkar, V. (eds) Languages and Compilers for Parallel Computing. LCPC 2010. Lecture Notes in Computer Science, vol 6548. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-19595-2_9
Download citation
DOI: https://doi.org/10.1007/978-3-642-19595-2_9
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-19594-5
Online ISBN: 978-3-642-19595-2
eBook Packages: Computer ScienceComputer Science (R0)