A performance modeling and optimizationanalysis tool for sparse matrix-vectormultiplication on GPUs

IEEE Transactions on Parallel and Distributed Systems

Volumn 25, Issue 5, 2014, Pages 1112-1123

  Guo, Ping ^a   Wang, Liqiang ^a   Chen, Po ^a  

^a UNIVERSITY OF WYOMING   (United States)

Author keywords

CUDA; GPU; Performance modeling; sparse matrix vector multiplication

Indexed keywords

EXPERIMENTS; OPTIMIZATION; PROGRAM PROCESSORS; TOOLS;

AVERAGE DIFFERENCE; CUDA; GPU; OPTIMAL SOLUTIONS; PERFORMANCE MODEL; SPARSE MATRICES; SPARSE MATRIX-VECTOR MULTIPLICATION; STORAGE FORMATS;

OPTIMAL SYSTEMS;

EID: 84898682038     PISSN: 10459219     EISSN: None     Source Type: Journal    
DOI: 10.1109/TPDS.2013.123     Document Type: Article

References (27)

1. N. Bell and M. Garland, "Implementing Sparse Matrix- VectorMultiplication On Throughput-Oriented Processors," Proc. Conf.High Performance Computing Networking, Storage and Analysis (SC'09), pp. 1-11, 2009.
2. A. Resios and V. Holdermans, "GPU Performance PredictionUsing Parametrized Models,", Master's thesis, Utrecht Univ.,2011.
3. P. Guo and L. Wang, "Accurate CUDA Performance Modeling forSparse Matrix-Vector Multiplication," Proc. IEEE Int'l Conf. HighPerformance Computing and Simulation (HPCS '12), pp. 496-502,July 2012.
4. P. Guo, H. Huang, Q. Chen, L. Wang, E.-J. Lee, and P. Chen,"A Model-Driven Partitioning and Auto-Tuning IntegratedFramework for Sparse Matrix-Vector Multiplication on GPUs,"Proc. TeraGrid Conf. Extreme Digital Discovery (TG '11), pp. 2:1-2:8, 2011.
5. J. Bolz, I. Farmer, E. Grinspun, and P. Schroder, "Sparse MatrixSolvers on The GPU: Conjugate Gradients and Multigrid," ACMTrans. Graphics, vol. 22, no. 3, pp. 917-924, 2003.
6. NVIDIA CUDA C Programming Guide, Version 4.0, May 2011.
7. J. Kurzak, W. Alvaro, and J. Dongarra, "Optimizing Matrix Multiplicationfor a Short-Vector Simd Architecture-Cell Processor,"J. Parallel Computing, vol. 35, no. 3, pp. 138-150, 2009.
8. E.-J. Im, K. Yelick, and R. Vuduc, "Sparsity: Optimization Frameworkfor Sparse Matrix Kernels," Int'l J. High Performance ComputingApplications, vol. 18, no. 1, pp. 135-158, 2004.
9. M.M. Baskaran and R. Bordawekar, "Optimizing Sparse Matrix-Vector Multiplication on GPUs," Research Report RC24704, IBMTJ Watson Research Center, Dec. 2008.
10. J. Demmel, J. Dongarra, V. Eijkhout, E. Fuentes, A. Petitet,R.C.W.R. Vuduc, and K. Yelick, "Self-Adapting Linear AlgebraAlgorithms and Software," Proc. IEEE, vol. 93, no. 2, pp. 293-312,Feb. 2005.
11. P. Guo and L. Wang, "Auto-Tuning CUDA Parameters forSparse Matrix-Vector Multiplication on GPUs," Proc. Int'l Conf.Computational and Information Sciences (ICCIS '10), pp. 1154-1157, 2010.
12. F. Vazquez, G. Ortega, J.J. Fernandez, and E.M. Garzon,"Improving the Performance of the Sparse Matrix Vector Productwith GPUs," Proc. 10th IEEE Int'l Conf. Computer and InformationTechnology (CIT '10), pp. 1146-1151, 2010.
13. A. Monakov, A. Lokhmotov, and A. Avetisyan, "AutomaticallyTuning Sparse Matrix-Vector Multiplication for GPUArchitectures," Proc. Fifth Int'l Conf. High Performance EmbeddedArchitectures and Compilers (HiPEAC '10), pp. 111-125, 2010.
14. D. Grewe and A. Lokhmotov, "Automatically Generating andTuning Gpu Code for Sparse Matrix-Vector Multiplication from aHigh-Level Representation," Proc. ACM Fourth Workshop GeneralPurpose Processing on Graphics Processing Units (GPGPU-4),pp. 12:1-12:8, 2011.
15. Z. Wang, X. Xu, W. Zhao, Y. Zhang, and S. He, "OptimizingSparse Matrix-Vector Multiplication on CUDA," Proc. Second Int'lConf. Education Technology and Computer (ICETC '10), vol. 4,pp. V4-109-V4-113, June 2010.
16. J.C. Pichel, F.F. Rivera, M. Fernandez, and A. Rodriguez, "Optimization of Sparse Matrix-Vector Multiplication Using ReorderingTechniques on GPUs," Microprocessors and Microsystems,vol. 36, no. 2, pp. 65-77, 2012.
17. X. Yang, S. Parthasarathy, and P. Sadayappan, "Fast SparseMatrix-Vector Multiplication on GPUs: Implications for GraphMining," Proc. VLDB Endowment, vol. 4, no. 4, pp. 231-242, Jan.2011.
18. S. Ryoo, C.I. Rodrigues, S.S. Stone, S.S. Baghsorkhi, S.-Z. Ueng,J.A. Stratton, and W.-M.W. Hwu, "Program Optimization SpacePruning for a Multithreaded GPU," Proc. ACM Sixth Ann. IEEE/ACM Int'l Symp. Code Generation and Optimization (CGO '08),pp. 195-204, 2008. (Pubitemid 351667266)
19. J.W. Choi, A. Singh, and R.W. Vuduc, "Model-Driven Autotuningof Sparse Matrix-Vector Multiply on GPUs," Proc. 15thACM SIGPLAN Symp. Principles and Practice of Parallel Programming(PPoPP '10), pp. 115-126, 2010.
20. D. Schaa and D. Kaeli, "Exploring the multiple-GPU DesignSpace," Proc. IEEE Int'l Parallel & Distributed Processing Symp.(IPDPS '09), pp. 1-12, May 2009.
21. S. Xu, W. Xue, and H. Lin, "Performance Modeling and Optimizationof Sparse Matrix-Vector Multiplication on NVIDIA CUDAPlatform," J. Supercomputing, vol. 63, pp. 710-721, 2013.
22. Y. Zhang and J. Owens, "A Quantitative Performance AnalysisModel for GPU Architectures," Proc. IEEE 17th Int'l Symp. HighPerformance Computer Architecture (HPCA '11), pp. 382-393, Feb.2011.
23. S.S. Baghsorkhi, M. Delahaye, S.J. Patel, W.D. Gropp, and W.-M.W. Hwu, "An Adaptive Performance Modeling Tool for GPUArchitectures," Proc. 15th ACM SIGPLAN Symp. Principles andPractice of Parallel Programming (PPoPP '10), pp. 105-114, 2010.
24. S. Hong and H. Kim, "An Analytical Model for a GPU Architecturewith Memory-Level and Thread-Level ParallelismAwareness," Proc. 36th ACM Ann. Int'l Symp. Computer Architecture(ISCA '09), pp. 152-163, 2009.
25. K. Kothapalli, R. Mukherjee, M. Rehman, S. Patidar, P. Narayanan,and K. Srinathan, "A Performance Prediction Model forthe CUDA GPGPU Platform," Proc. Int'l Conf. High PerformanceComputing (HiPC '09), pp. 463-472, Dec. 2009.
26. T.A. Davis and Y. Hu, "The University of Florida Sparse MatrixCollection," ACM Trans. Math. Software, vol. 38, no. 1, pp. 1:1-1:25,2011.
27. S. Williams, L. Oliker, R. Vuduc, J. Shalf, K. Yelick, and J. Demmel,"Optimization of Sparse Matrix-Vector Multiplication on EmergingMulticore Platforms," Proc. ACM/IEEE Conf. Supercomputing,2007.