Parallelizing dense and banded linear algebra libraries using SMPSs

Concurrency and Computation: Practice and Experience

Volumn 21, Issue 18, 2009, Pages 2438-2456

  Badia, Rosa M ^a,b   Herrero, José R ^c   Labarta, Jesús ^a   Pérez, Josep M ^a   Quintana Ortí, Enrique S ^d   Quintana Ortí, Gregorio ^d  

^a BARCELONA SUPERCOMPUTING CENTER   (Spain)

^b CSIC   (Spain)

^c UNIVERSITAT POLITÈCNICA DE CATALUNYA   (Spain)

^d UNIVERSITAT JAUME I   (Spain)

Author keywords

Dynamic scheduling; High performance; Linear algebra libraries; Multi core processors; Programmability

Indexed keywords

APPLICATION PROGRAMMING INTERFACES (API); DIGITAL STORAGE; LINEAR ALGEBRA;

DYNAMIC SCHEDULING; HIGH PERFORMANCE; LINEAR ALGEBRA LIBRARIES; MULTI-CORE PROCESSOR; PROGRAMMABILITY;

LIBRARIES;

EID: 73349095700     PISSN: 15320626     EISSN: 15320634     Source Type: Journal    
DOI: 10.1002/cpe.1463     Document Type: Article

References (28)

1. Anderson E, Bai Z, Demmel J, Dongarra JE, DuCroz J, Greenbaum A, Hammarling S, McKenney AE, Ostrouchov S, Sorensen D. LAPACK Users' Guide. SIAM: Philadelphia, 1992.
2. Buttari A, Langou J, Kurzak J, Dongarra J. Parallel tiled QR factorization for multicore architectures. Concurrency and Computation: Practice and Experience 2008; 20(13): 1573-1590.
3. Buttari A, Langou J, Kurzak J, Dongarra J. A class of parallel tiled linear algebra algorithms for multicore architectures. Parallel Computing 2009; 35(1): 38-53.
4. Chan E, Quintana-Ortí ES, Quintana-Ortí G, van de Geijn R. Super matrix out-of-order scheduling of matrix operations for SMP and multi-core architectures. Proceedings of the Nineteenth ACM Symposium on Parallelism in Algorithms and Architectures (SPAA 2007), San Diego, CA, U.S.A., 9-11 June 2007; 116-125.
5. Chan E, Van Zee FG, Quintana-Ortí ES, Quintana-Ortí G, van de Geijn R. Satisfying your dependencies with super matrix. Proceedings of IEEE Cluster Computing 2007, September 2007; 91-99.
6. Quintana-Ortí G, Quintana-Ortí ES, Chan E, van de Geijn R, Van Zee FG. Design of scalable dense linear algebra libraries for multithreaded architectures: the LU factorization. Workshop on Multithreaded Architectures and Applications - MTAAP 2008, 2008. CD-ROM.
7. Quintana-Ortí G, Quintana-Ortí ES, Chan E, Van Zee FG, van de Geijn RA. Scheduling of QR factorization algorithms on SMP and multi-core architectures. 16th Euromicro International Conference on Parallel, Distributed and Network-based Processing - PDP 2008, El Baz FSD, Bourgeois J (eds.). 2008; 301-310.
8. Chan E, Van Zee FG, Bientinesi P, Quintana-Ortí ES, Quintana-Ortí G, van de Geijn R. Super matrix: A multithreaded runtime scheduling system for algorithms-by-blocks. ACM SIGPLAN 2008 Symposium on Principles and Practices of Parallel Programming (PPoPP'08), 2008; 123-132.
9. Quintana-Ortí G, Quintana-Ortí ES,Remón A, van de Geijn R. Supermatrix for the factorization of band matrices. FLAME Working Note #27 TR-07-51, The University of Texas at Austin, Department of Computer Sciences, September 2007.
10. Blumofe RD, Joerg CF, Kuszmaul BC, Leiserson CE, Randall KH, Zhou Y. Cilk: An efficient multithreaded runtime system. Journal of Parallel and Distributed Computing 1996; 37(1): 55-69.
11. Bellens P, Pérez JM, Badia RM, Labarta J. CellSs: A programming model for the Cell BE architecture. SC '06: Proceedings of the 2006 ACM/IEEE Conference on Supercomputing. ACM Press: New York, NY, U.S.A., 2006; 86.
12. Pérez JM, Bellens P, Badia RM, Labarta J. CellSs programming the Cell/B.E. made easier. IBM Journal of Research and Development 2007; 51(5).
13. Pérez JM, Badia RM, Labarta J. A flexible and portable programming model for SMP and multi-cores. Technical Report 03/2007, Barcelona Supercomputing Center - Centro Nacional de Supercomputacion, Barcelona, Spain, 2007.
14. Pérez JM, Badia RM, Labarta J. A dependency-aware task-based programming environment for multi-core architectures. Proceedings of the 2008 IEEE International Conference on Cluster Computing, Causal Productions (ed.). September 2008; 142-151. IEEE Catalog Number CFP08235-CDR.
15. Golub GH, Van Loan CF. Matrix Computations (3rd edn). The Johns Hopkins University Press: Baltimore, MD, 1996.
16. Gunnels JA, Gustavson FG, Henry GM, van de Geijn RA. FLAME: Formal linear algebra methods environment. ACM Transactions on Mathematical Software 2001; 27(4): 422-455.
17. Bientinesi P, Gunnels JA, Myers ME, Quintana-Ortí ES, van de Geijn RA. The science of deriving dense linear algebra algorithms. ACM Transactions on Mathematical Software 2005; 31(1): 1-26.
18. Bientinesi P, Quintana-Ortí ES, van de Geijn RA. Representing linear algebra algorithms in code: The FLAME application programming interfaces. ACM Transactions on Mathematical Software 2005; 31(1): 27-59.
19. Strazdins P. A comparison of lookahead and algorithmic blocking techniques for parallel matrix factorization. Technical Report TR-CS-98-07, Department of Computer Science, The Australian National University, Canberra 0200 ACT, Australia, 1998.
20. Kågström B, Ling P, Loan CV. GEMM-based level 3 BLAS: High-performance model implementations and performance evaluation benchmark. ACM Transactions on Mathematical Software 1998; 24(3): 268-302.
21. Kågström B, Ling P, Loan CV. Algorithm 784: GEMM-based level 3 BLAS: portability and optimization issues. ACM Transactions on Mathematical Software 1998; 24(3): 303-316.
22. Gunter BC, van de Geijn RA. Parallel out-of-core computation and updating the QR factorization. ACM Transactions on Mathematical Software 2005; 31(1): 60-78.
23. Joffrain T, Quintana-Ortí ES, van de Geijn RA. Rapid development of high-performance out-of-core solvers. Proceedings of PARA 2004, Lecture Notes in Computer Science, vol. 3732. Springer: Berlin, Heidelberg, 2005 ; 413-422.
24. Quintana-Ortí ES, van de Geijn R. Updating an LU factorization with pivoting. ACM Transactions on Mathematical Software 2008; 35(2): 11: 1-11: 16.
25. Gustavson FG. New generalized matrix data structures lead to a variety of high-performance algorithms. The Architecture of Scientific Software, Boisvert RF, Tang PTP (eds.), vol. 188 of IFIP Conference Proceedings. Kluwer: Dordrecht, 2000; 211-234.
26. Park N, Hong B, Prasanna VK. Tiling, block data layout, and memory hierarchy performance. IEEE Transactions on Parallel and Distributed Systems 2003; 14(7): 640-654.
27. Herrero JR. A framework for efficient execution of matrix computations. PhD Thesis, Polytechnic University of Catalonia, Spain, 2006.
28. Low TM, van de Gejin R. An API for manipulating matrices stored by blocks. Technical Report TR-2004-15, Department of Computer Sciences, The University of Texas at Austin, May 2004.