Abstract
Graphics processing units (GPUs) have shown increased popularity in general-purpose parallel processing. This massively parallel architecture allows GPUs to execute tens of thousands of threads in parallel to solve heavily data-parallel problems efficiently. However, despite the tremendous computing power, optimizing GPU kernels to achieve high performance is still a challenge due to the sea change from CPU to GPU and lacking of tools for programming and performance analysis.
In this paper, we propose a throughput-aware analytical model to estimate the performance of GPU kernels and optimizations. We construct a pipeline for global memory access servicing and redefine the compute throughput and memory throughput as the speed of memory requests arriving and leaving the pipeline. Based on concluding the kernel throughput limiting factor, GPU programs are classified into compute-bound and memory-bound categories and then we predict performance for each category. Besides, our model can provide useful information on the direction of optimization and predict the potential performance benefits. We demonstrate our model on a manually written benchmark as well as the matrix-multiply kernel and show that the geometric mean of absolute error of our model is less than 6.5%.
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
Keckler, S.W., Dally, W.J., Khailany, B., et al.: GPUs and the future of parallel computing. IEEE Micro 31(5), 7–17 (2011)
Advanced Micro Devices, Inc. AMD Brook+
NVIDIA Corporation. CUDA Programming Guide, Version 4.0
Stone, J.E., Gohara, D., Shi, G.: OpenCL: A parallel programming standard for heterogeneous computing systems. Computing in Science & Engineering 12(3), 66 (2010)
Owens, J.D., Houston, M., Luebke, D., et al.: GPU computing. Proceedings of the IEEE 96(5), 879–899 (2008)
Lindholm, E., Nickolls, J., Oberman, S., et al.: NVIDIA Tesla: A unified graphics and computing architecture. IEEE Micro 28(2), 39–55 (2008)
Williams, S., Waterman, A., Patterson, D.: Roofline: an insightful visual performance model for multicore architectures. Communications of the ACM 52(4), 65–76 (2009)
Zhang, Y., Owens, J.D.: A quantitative performance analysis model for GPU architectures. In: 2011 IEEE 17th International Symposium on High Performance Computer Architecture (HPCA), pp. 382–393. IEEE (2011)
Hong, S., Kim, H.: An analytical model for a GPU architecture with memory-level and thread-level parallelism awareness. ACM SIGARCH Computer Architecture News 37(3), 152–163 (2009)
Baghsorkhi, S.S., Delahaye, M., Patel, S.J., et al.: An adaptive performance modeling tool for GPU architectures. ACM Sigplan Notices 45(5), 105–114 (2010)
Meng, J., Morozov, V.A., Kumaran, K., et al.: GROPHECY: GPU performance projection from CPU code skeletons. In: Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis, p. 14. ACM (2011)
Cui, Z., et al.: An accurate GPU performance model for effective control flow divergence optimization. In: 2012 IEEE 26th International Parallel & Distributed Processing Symposium (IPDPS). IEEE (2012)
Che, S., Boyer, M., Meng, J., et al.: Rodinia: A benchmark suite for heterogeneous computing. In: IEEE International Symposium on Workload Characterization, IISWC 2009, pp. 44–54. IEEE (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Hu, Z., Liu, G., Dong, W. (2014). A Throughput-Aware Analytical Performance Model for GPU Applications. In: Wu, J., Chen, H., Wang, X. (eds) Advanced Computer Architecture. Communications in Computer and Information Science, vol 451. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-662-44491-7_8
Download citation
DOI: https://doi.org/10.1007/978-3-662-44491-7_8
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-662-44490-0
Online ISBN: 978-3-662-44491-7
eBook Packages: Computer ScienceComputer Science (R0)