High performance matrix inversion on a multi-core platform with several GPUs

Proceedings - 19th International Euromicro Conference on Parallel, Distributed, and Network-Based Processing, PDP 2011

Volumn , Issue , 2011, Pages 87-93

  Ezzatti, Pablo ^a   Quintana Orti, Enrique S ^b   Remon, Alfredo ^b  

^a UNIVERSIDAD DE LA REPÚBLICA   (Uruguay)

^b UNIVERSITAT JAUME I   (Spain)

Author keywords

GPUs; linear algebra; matrix inversion

Indexed keywords

COMPUTATIONAL EFFORT; COMPUTATIONAL KERNELS; COMPUTATIONAL UNITS; CONCURRENT EXECUTION; CORE PROCESSORS; FLOATING POINT OPERATIONS PER SECONDS; GAUSS-JORDAN ELIMINATION; GENERAL PURPOSE PROCESSORS; GPUS; GRAPHICS PROCESSOR; HIGH PERFORMANCE CODES; HIGH PERFORMANCE COMPUTING; MATRIX INVERSIONS; MODEL REDUCTION; MULTI-CORE PLATFORMS; MULTI-CORE PROCESSOR; NUMERICAL EXPERIMENTS; OFF-LOAD; OPTIMAL CONTROLS; PARALLEL IMPLEMENTATIONS; PERFORMANCE MATRICES; SCIENTIFIC APPLICATIONS; TARGET ARCHITECTURES;

ALGEBRA; COMPUTER ARCHITECTURE; COMPUTER SOFTWARE SELECTION AND EVALUATION; GENERAL PURPOSE COMPUTERS; NETWORK ARCHITECTURE; PARALLEL PROCESSING SYSTEMS; PROGRAM PROCESSORS;

MATRIX ALGEBRA;

EID: 79955025807     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/PDP.2011.66     Document Type: Conference Paper

References (14)

1. V. Volkov, J. Demmel, "Benchmarking GPUs to tune dense linear algebra," SC '08 Proceedings of the 2008 ACM/IEEE conference on Supercomputing, 2008.
2. "Exploiting the capabilities of modern GPUs for dense matrix computations," Concurrency and Computation: Practice and Experience, vol. 21, pp. 2457-2477, 2009.
3. "Using hybrid CPU-GPU platforms to accelerate the computation of the matrix sign function," in Lecture Notes in Computer Science, 7th Int. Workshop on Algorithms, Models and Tools for Parallel Computing on Heterogeneous Networks - HeteroPar'09, 2009.
4. A. V. Gerbessiotis, "Algorithmic and practical considerations for dense matrix computations on the BSP model," Oxford University Computing Laboratory, PRG-TR 32, 1997. [Online]. Available: http://web.njit.edu/~alexg/ pubs/papers/PRG3297.ps
5. "A note on parallel matrix inversion," SIAM J. Sci. Comput., vol. 22, pp. 1762-1771, 2001.
6. J. A. Gunnels, F. G. Gustavson, G. M. Henry, and R. A. van de Geijn, "FLAME: Formal linear algebra methods environment," ACM Trans. Math. Soft., vol. 27, no. 4, pp. 422-455, December 2001. [Online]. Available: http://doi.acm.org/10.1145/504210.504213 (Pubitemid 33602331)
7. "The science of deriving dense linear algebra algorithms," ACM Trans. Math. Soft., vol. 31, no. 1, pp. 1-26, March 2005. [Online]. Available: http://doi.acm.org/10.1145/1055531.1055532 (Pubitemid 40557860)
8. A. Strazdins, "A comparison of lookahead and algorithmic blocking techniques for parallel matrix factorization," Tech. Rep. TR-CS-98-07, Department of Computer Science, The Australian National University, Canberra 0200 ACT, Australia 1998.
9. "Solving dense linear algebra problems on platforms with multiple hardware accelerators," ACM SIGPLAN Symposium on Principles and Practice of Parallel Computing, pp. 121-129, 2009.
10. "Updating an LU factorization with pivoting," ACM Trans. Math. Softw., vol. 35, no. 2, pp. 1-16, 2008. [Online]. Available: http://doi.acm.org/10.1145/1377612.1377615
11. "An extension of the StarSs programming model for platforms with multiple GPUs," Lecture Notes in Computer Science 5704, Euro-Par 2009, pp. 851-862, 2009.
12. "Extending OpenMP to survive the heterogeneous multi-core era," Int. Journal of Parallel Programming, vol. 38, no. 5, pp. 440-459, 2010. [Online]. Available: http://doi.acm.org/10.1145/1377612.1377615
13. S. Tomov, R. Nath, H. Ltaief, J. Dongarra, "Dense linear algebra solvers for multicore with GPU accelerators," Published at HIPS 2010, Atlanta, GA, January 2010.
14. "StarPU: A unified platform for task scheduling on heterogeneous multicore architectures," Lecture Notes in Computer Science 5704, Euro-Par 2009, pp. 863-874, 2009.