A predictive model for solving small linear algebra problems in GPU registers

Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium, IPDPS 2012

Volumn , Issue , 2012, Pages 2-13

  Anderson, Michael J ^a   Sheffield, David ^a   Keutzer, Kurt ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

Dense Linear Algebra; GPGPU; Modeling

Indexed keywords

ADAPTIVE RADAR; COMPLEX MATRICES; DENSE LINEAR ALGEBRA; GPGPU; GRAPHICS PROCESSING UNITS; HARDWARE-ACCELERATED; KERNEL LIBRARIES; LINEAR ALGEBRA PROBLEMS; MATRIX SIZE; MEMORY HIERARCHY; MEMORY SUBSYSTEMS; MULTI-LEVEL PARALLELISM; NUMERICAL LINEAR ALGEBRA; PREDICTIVE MODELS; PROBLEM SIZE; QR FACTORIZATIONS; RADAR PROCESSING; REAL-TIME APPLICATION; SQUARE ROOT FUNCTIONS; SUPERCOMPUTING APPLICATIONS;

COMPUTER GRAPHICS; DISTRIBUTED PARAMETER NETWORKS; FACTORIZATION; LINEAR ALGEBRA; MEMORY ARCHITECTURE; MODELS; PROGRAM PROCESSORS; SIGNAL PROCESSING;

PROBLEM SOLVING;

EID: 84866859911     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/IPDPS.2012.11     Document Type: Conference Paper

References (23)

1. CUDA forums: Lots of small matrices. http://forums.nvidia.com/index.php? showtopic=188430. Accessed: 09/26/2011.
2. CULA - a hybrid GPU linear algebra package. http://nvidia.fullviewmedia. com/gtc2010/0923-a3-2153.html. NVIDIA GPU Technology Conference 2010.
3. CULA forums: Batch level parallelism. http://www.culatools.com/forums/ viewtopic.php?f=14&t=774. Accessed: 09/26/2011.
4. Magma forums: parallel execution of multiple magma dsyevd. http://icl.cs.utk.edu/magma/forum/viewtopic.php?f=2&t=204&p= 848&hilit=streams#p848. Accessed: 09/26/2011.
5. E. Agullo, J. Demmel, J. Dongarra, B. Hadri, J. Kurzak, J. Langou, H. Ltaief, P. Luszczek, and S. Tomov. Numerical linear algebra on emerging architectures: The plasma and magma projects. In Journal of Physics: Conference Series, volume 180, page 012037. IOP Publishing, 2009.
6. G. Ballard, J. Demmel, O. Holtz, and O. Schwartz. Minimizing communication in linear algebra. Arxiv preprint arXiv:0905.2485, 2009.
7. P.B. Bhat, Y.W. Lim, and V.K. Prasanna. Issues in using heterogeneous hpc systems for embedded real time signal processing applications. In rtcsa, page 134. Published by the IEEE Computer Society, 1995.
8. K.C. Cain. Rt-stap: Real-time space-time adaptive processing benchmark. Technical report, DTIC Document, 1997.
9. S. Chatterji, M. Narayanan, J. Duell, and L. Oliker. Performance evaluation of two emerging media processors: VIRAM and Imagine. 2003.
10. D. Culler, R. Karp, D. Patterson, A. Sahay, K.E. Schauser, E. Santos, R. Subramonian, and T. Von Eicken. LogP: Towards a realistic model of parallel computation, volume 28. ACM, 1993.
11. J.W. Demmel et al. Applied numerical linear algebra, volume 150. SIAM Philadelphia, PA,, USA, 1997.
12. P.R. Dixon, T. Oonishi, and S. Furui. Harnessing graphics processors for the fast computation of acoustic likelihoods in speech recognition. Computer Speech & Language, 23(4):510-526, 2009.
13. J.J. Dongarra, J. Du Croz, S. Hammarling, and I.S. Duff. A set of level 3 basic linear algebra subprograms. ACM TOMS, 16(1):1-17, 1990.
14. J.R. Humphrey, D.K. Price, K.E. Spagnoli, A.L. Paolini, and E.J. Kelmelis. CULA: hybrid GPU accelerated linear algebra routines. In SPIE Conference Series, volume 7705, page 1, 2010.
15. Wei keng Liao, A. Choudhary, D. Weiner, and P. Varshney. Multi-threaded design and implementation of parallel pipelined stap on parallel computers with smp nodes. pages 448 -452, apr 1999.
16. M. Murphy, K. Keutzer, S. Vasanawala, and M. Lustig. Clinically feasible reconstruction time for l1-spirit parallel imaging and compressed sensing mri. ISMRM'10, 2010.
17. R. Nath, S. Tomov, and J. Dongarra. An improved magma GEMM for Fermi GPUs. ICL, University of Tennessee, Tech. Rep, 2010.
18. J. Nickolls and W.J. Dally. The GPU computing era. Micro, IEEE, 30(2):56-69, 2010.
19. Nvidia. CUDA programming guide. 2009.
20. Michael Parker. Radar basics. http://www.eetimes.com/ design/programmable-logic/4216104/Radar-basics-Part-1. Accessed: 09/27/2011.
21. V. Volkov and J.W. Demmel. Benchmarking GPUs to tune dense linear algebra. In High Performance Computing, Networking, Storage and Analysis, 2008. SC 2008. International Conference for, pages 1-11, Nov. 2008.
22. S. Williams, A. Waterman, and D. Patterson. Roofline: an insightful visual performance model for multicore architectures. Communications of the ACM, 52(4):65-76, 2009.
23. H. Wong, M.-M. Papadopoulou, M. Sadooghi-Alvandi, and A. Moshovos. Demystifying GPU microarchitecture through microbenchmarking. In ISPASS, pages 235-246, March 2010.