New Row-grouped CSR format for storing sparse matrices on GPU with implementation in CUDA

Acta Technica CSAV (Ceskoslovensk Akademie Ved)

Volumn 56, Issue 4, 2011, Pages 447-466

  Oberhuber, Tomáš ^a   Suzuki, Atsushi ^b   Vacata, Jan ^a  

^a CZECH TECHNICAL UNIVERSITY IN PRAGUE   (Czech Republic)

^b UNIVERSITÉ PIERRE ET MARIE CURIE   (France)

Author keywords

CUDA; GPU; Parallel computing; Sparse matrices; SpMV; Thread computing

Indexed keywords

CUDA; GPU; SPARSE MATRICES; SPMV; THREAD COMPUTING;

ELECTRICAL ENGINEERING; MECHANICAL PROPERTIES; PARALLEL PROCESSING SYSTEMS;

PARALLEL ARCHITECTURES;

EID: 84857836454     PISSN: 00017043     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (20)

1. V. W. Lee, C. Kim, J. Chhugani, M. Deisher, D. Kim, A. D. Nguyen, N. Satish, M. Smelyanskiy, S. Chennupaty, P. Hammarlund, R. Singhal, P. Dubey: Debunking the 100X GPU vs. CPU myth: An evaluation of throughput computing on CPU and GPU. Proc. 37th Ann. Int. Symposium on Computer Architecture (ISCA'10) Saint-Malo (France), June 19-23, 2010, ACM, New York 2010, 451-460.
2. NVIDIA Corporation, CUDA CUBLAS library, PG-00000-002 V3.1, May 2010, http://developer.download.nvidia.com/compute/cuda/3-1/toolkit/docs/ CUBLAS-Library-3.1.pdf.
3. N. Galoppo, N. K Govindaraju, M. Henson, D. Manocha: LU-GPU: Efficient algorithms for solving dense linear systems on graphics hardware. Proceedings ACM/ IEEE SC'05, Conference of Supercomputing, Nov. 12-18, 2005, Seattle (USA), doi: 10.1109/SC.2005.42. (Pubitemid 44902346)
4. V. Volkov, J. Demel: LU, QL and Cholesky factorizations using vector capabilities of GPUs. Techn. Rep. UCB/EECS-2008-49, Electrical Engineering and Computer Sciences, University of California, Berkeley, 2008.
5. N. Bell, M. Garland: Efficient sparse matrix-vector multiplication on CUDA. Techn. Rep. NVR-2008-004, NVIDIA Corporation 2008.
6. E.-J. Im: Optimizing the performance of sparse matrix-vector multiplication. PhD thesis, Rep. UCB/CSD-00-1104, University of California, Berkeley, 2000.
7. K. Kourtis, G. Goumas, N. Koziris: Improving the performance of multithreaded sparse matrix-vector multiplication using index and value compression. Proc. 37th International Conference on Parallel Processing, Sept. 8-12, 2008, Portland (USA), 511-519.
8. M. M. Baskaran, R. Bordwaker: Optimizing sparse matrix-vector multiplication on GPUs. Techn. Rep. RC24704(W0812-047), IBM 2008, http://domino.watson.ibm.com/library/CyberDig.nsf/papers/ 1D32F6D23B99F7898525752200618339/$File/rc24704.pdf.
9. J. W. Choi, A. Singh, R. Vuduc: Model-driven autotuning of sparse matrix-vector multipy on GPUs. Proc. 15th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP 2010), Bangalore (India), Jan. 9-14, 2010 (R. Govindarajan, D. A. Padua, M. W. Hall, eds.), ACM 2010, 37-48.
10. A. Monakov, A. Avetisyan: Implementing blocked sparse matrix-vector multiplication on NVIDIA GPUs. Proc. 9th International Workshop on Embedded Computer Systems: Architectures, Modeling, and Simulation, Samos (Greece), July 20-23, 2009 (K. Bertels, N. J. Dimopoulos, C. Silvano, S. Wong, eds.) Springer, Berlin 2009, 289-297.
11. T. A. Davis, Y. Hu: The University of Florida sparse matrix collection. NA Digest 92 (42), http://www.cise.ufl.edu/research/sparse/matrices/.
12. Z. Bai, D. Day, J. Demmel, J. Dongarra: Test matrix collection (non-Hermitian eigenvalue problems). release 1, Techn. Rep., University of Kentucky, 1996, http://math.nist.gov/MatrixMarket/
13. NVIDIA CUDA Programming Guide 3.0. NVIDIA Corporation, 2010, http://developer.download.nvidia.com/compute/cuda/3-0/toolkit/docs/ NVIDIA-CUDA-ProgrammingGuide.pdf.
14. L. Buatois, G. Caumon, B. Levy: Concurrent number cruncher: a gpu implementation of a general sparse linear solver. Int. J. Parallel Emerg. Distrib. Syst. 24 (2009), 205-223.
15. Amd "close to metal" technology unleashes the power of stream computing. Techn. Rep., AMD Press Release 2006.
16. J. Vacata: GPGPU: General purpose computation on GPUs. Master's thesis, Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, 2008.
17. A. Monakov, A. Lokhmotov, A. Avetisyan: Automatically tuning sparse matrixvector multiplication for GPU architectures. Proc. 5th International Conferences on High Performance Embedded Architectures and Compilers (HiPEAC 2010), Pisa (Italy), Jan. 25-27, 2010 (Y. N. Patt, P. Foglia, E. Duesterwald, P. Faraboschi, X. Martorell, eds.), Springer, Berlin 2010, 111-125.
18. Nvidia, Cusp 0.1.1. http://code.google.com/p/cusp-library/, 2010.
19. R. Nishtala, R. W. Vuduc, J. W. Demmel, K. A. Yelick: When cache blocking of sparse matrix vector multiply works and why. Appl. Algebra Eng. Commun. Comput. 18 (2007), 297-311. (Pubitemid 47224626)
20. P. Amestoy, T. A. Davis, I. S. Duff: An approximate minimum degree ordering algorithm. SIAM J. Matrix Anal. Appl. 17 (1996), 886-905.