Advertisement

ATLAS Version 3.9: Overview and Status

  • R. Clint Whaley
Chapter

Abstract

This paper describes the widely used ATLAS (Automatically Tuned Linear Algebra Software) project as it stands today. ATLAS is an instantiation of a paradigm in high performance library production and maintenance, which we term AEOS (Automated Empirical Optimization of Software); this style of library management has been created to allow software to keep pace with the incredible rate of hardware advancement inherent in Moore’s Law. ATLAS is the application of this AEOS paradigm to dense linear algebra software. ATLAS produces a full BLAS (Basic Linear Algebra Subprograms) library as well as provides some optimized routines for LAPACK (Linear Algebra PACKage). This paper overviews the basics of what ATLAS is and how it works, highlights some of the recent improvements available as of version 3.9.23, in addition to discussing some of the current challenges and future work.

Keywords

Cache State Cache Utilization Multiple Implementation Linear Algebra PACKage Panel Factorization 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Andersen BS, Gustavson FG, Wasniewski J (2000) A recursive formulation of cholesky factorization of a matrix in packed storage. Technical Report UT CS-00-448, LAPACK Working Note No. 146, University of TennesseeGoogle Scholar
  2. 2.
    Anderson E, Bai Z, Bischof C, Demmel J, Dongarra J, Du Croz J, Greenbaum A, Hammarling S, McKenney A, Ostrouchov S, Sorensen D (1999) LAPACK users’ guide, 3rd edn. SIAM, Philadelphia, PACrossRefGoogle Scholar
  3. 3.
    Baradaran N, Chame J, Chen C, Diniz P, Hall M, Lee Y-J, Liu B, Lucas R (2003) Eco: An empirical-based compilation and optimization system. In: International parallel and distributed processing symposium, 2003Google Scholar
  4. 4.
    Bilmes J, Asanović K, Chin CW, Demmel J (1997) Optimizing matrix multiply using PHiPAC: a portable, high-performance, ANSI C coding methodology. In: Proceedings of the ACM SIGARC international conference on superComputing, Vienna, Austria, July 1997Google Scholar
  5. 5.
    Blackford LS, Demmel J, Dongarra J, Duff I, Hammarling S, Henry G, Heroux M, Kaufman L, Lumsdaine A, Petitet A, Pozo R, Remington K, Whaley RC (2002) An updated set of basic linear algebra subprograms (BLAS). ACM Trans Math Software 28(2):135–151MathSciNetCrossRefGoogle Scholar
  6. 6.
    Blackford S, Corliss G, Demmel J, Dongarra J, Duff I, Hammarling S, Henry G, Heroux M, Hu C, Kahan W, Kaufman L, Kearfott B, Krogh F, Li X, Maany Z, Petitet A, Pozo R, Remington K, Walster W, Whaley C, Wolff J, Gudenberg V (1999) Document for the basic linear algebra subprograms (BLAS) standard: BLAS technical forum. http://www.netlib.org/cgi-bin/checkout/blast/blast.pl
  7. 7.
    Castaldo AM, Whaley RC (2009) Minimizing startup costs for performance-critical threading. In: Proceedings of the IEEE international parallel and distributed processing symposium, Rome, Italy, May 2009Google Scholar
  8. 8.
    Castaldo AM, Whaley RC (2010) Scaling LAPACK panel operations using parallel cache assignment. In: 15th ACM SIGPLAN annual symposium on principles and practice of parallel programming, Bangalore, India, Jan 2010Google Scholar
  9. 9.
    Dayde M, Duff I, Petitet A (1994) A parallel block implementation of level 3 BLAS for MIMD vector processors. ACM Trans Math Software 20(2):178–193MATHCrossRefGoogle Scholar
  10. 10.
    Diniz P, Lee Y-J, Hall M, Lucas R (2004) A case study using empirical optimization for a large, engineering application. In: International parallel and distributed processing symposium, 2004. CD-ROM ProceedingsGoogle Scholar
  11. 11.
    Dongarra J, Du Croz J, Duff I, Hammarling S (1990) A set of level 3 basic linear algebra subprograms. ACM Trans Math Software 16(1):1–17MATHCrossRefGoogle Scholar
  12. 12.
    Dongarra J, Du Croz J, Hammarling S, Hanson R (1988) Algorithm 656: an extended set of basic linear algebra subprograms: model implementation and test programs. ACM Trans Math Software 14(1):18–32MATHCrossRefGoogle Scholar
  13. 13.
    Dongarra J, Du Croz J, Hammarling S, Hanson R (1988) An extended set of FORTRAN basic linear algebra subprograms. ACM Trans Math Software 14(1):1–17MATHCrossRefGoogle Scholar
  14. 14.
    Elmroth E, Gustavson F (2000) Applying recursion to serial and parallel qr factorization leads to better performance. IBM J Res Dev 44(4):605–624CrossRefGoogle Scholar
  15. 15.
    Frigo M, Johnson S (1998) FFTW: an adaptive software architecture for the FFT. In: Proceedings of the international conference on acoustics, speech, and signal processing (ICASSP), vol 3, p 1381Google Scholar
  16. 16.
    Frigo M, Johnson SG (1997) The fastest fourier transform in the west. Technical Report MIT-LCS-TR-728, Massachusetts Institute of TechnologyGoogle Scholar
  17. 17.
    Gustavson F (1997) Recursion leads to automatic variable blocking for dense linear-algebra algorithms. IBM J Res Dev 41(6):737–755CrossRefGoogle Scholar
  18. 18.
    Gustavson F, Henriksson A, Jonsson I, Kågström B, Ling P (1998) Recursive blocked data formats and blas’s for dense linear algebra algorithms. In: Kågström B, Dongarra J, Elmroth E, Waśniewski J (eds) Applied parallel computing, PARA’98, Lecture notes in computer science, No. 1541, pp 195–206Google Scholar
  19. 19.
    Gustavson F, Henriksson A, Jonsson I, Kågström B, Ling P (1998) Superscalar gemm-based level 3 blas – the on-going evolution of a portable and high-performance library. In: Kågström B, Dongarra J, Elmroth E, Waśniewski J (eds) Applied parallel computing, PARA’98, Lecture Notes in Computer Science, No. 1541, pp 207–215Google Scholar
  20. 20.
    Hanson R, Krogh F, Lawson C (1973) A proposal for standard linear algebra subprograms. ACM SIGNUM Newsl 8(16)Google Scholar
  21. 21.
    Kågström B, Ling P, van Loan C (1995) GEMM-based level 3 BLAS: high-performance model implementations and performance evaluation benchmark. Technical Report UMINF 95-18, Department of Computing Science, Umeå UniversityGoogle Scholar
  22. 22.
    Kågström B, Ling P, van Loan C (1998) Gemm-based level 3 blas: High performance model implementations and performance evaluation benchmark. ACM Trans Math Software 24(3):268–302MATHCrossRefGoogle Scholar
  23. 23.
    Kågström B, Ling P, van Loan C (1998) Gemm-based level 3 blas: High performance model implementations and performance evaluation benchmark. ACM Trans Math Software 24(3):268–302MATHCrossRefGoogle Scholar
  24. 24.
    Kisuki T, Knijnenburg PMW, O’Boyle MFP, Bodin F, Wijshoff HAG (1999) A feasibility study in iterative compilation. In: ISHPC, pp 121–132Google Scholar
  25. 25.
    Lawson C, Hanson R, Kincaid D, Krogh F (1979) Basic linear algebra subprograms for fortran usage. ACM Trans Math Software 5(3):308–323MATHCrossRefGoogle Scholar
  26. 26.
    Moura JMF, Johnson J, Johnson RW, Padua D, Prasanna V, Pschel M, Veloso MM (1998) Spiral: automatic library generation and platform-adaptation for dsp algorithms. http://www.ece.cmu.edu/~spiral
  27. 27.
    Pouchet L-N, Bastoul C, Cohen A, Vasilache N (2007) Iterative optimization in the polyhedral model: Part i, one-dimensional time. In: Code generation and optimization, 2007. CGO ’07. international symposium on, pp 144–156, Mar 2007Google Scholar
  28. 28.
    Pushel M, Moura J, Johnson J, Padua D, Veloso M, Singer B, Xiong J, Frenchetti F, Cacic A, Voronenko Y, Chen K, Johnson R, Rizzolo N (2005) Spiral: code generation for dsp transforms. In: Proceedings of the IEEE, 93(2). special issue on “Program generation, optimization, and adaptation”Google Scholar
  29. 29.
    See page for details. FFTW homepage. http://www.fftw.org/
  30. 30.
    See page for details. SPIRAL homepage. http://www.spiral.net/
  31. 31.
    Tiwari A, Chen C, Chame J, Hall M, Hollingsworth JK (2009) A scalable autotuning framework for compiler optimization. In: Proceedings of the IEEE international parallel and distributed processing symposium, Rome, Italy, May 2009Google Scholar
  32. 32.
    Toledo S (1997) Locality of reference in lu decomposition with partial pivoting. SIAM J Matrix Anal Appl 18(4)Google Scholar
  33. 33.
    van der Mark P, Rohou E, Bodin F, Chamski Z, Eisenbeis C (1999) Using iterative compilation for managing software pipeline – unrolling tradoffs. In: SCOPES99, St. Goar, GermanyGoogle Scholar
  34. 34.
    Vuduc R, Demmel JW, Yelick KA (2005) OSKI: A library of automatically tuned sparse matrix kernels. In: Proceedings of SciDAC 2005, Journal of Physics: Conference Series, San Francisco, CA, USA, vol 16, pp 521–530Google Scholar
  35. 35.
    Whaley RC User contribution to atlas. http://math-atlas.sourceforge.net/devel/atlas_contrib/ also available in ATLAS/doc/atlas_contrib.pdf of tarfile
  36. 36.
    Whaley RC (2008) Empirically tuning lapack’s blocking factor for increased performance. In: Proceedings of the international multiconference on computer science and information technology, Wisla, Poland, October 2008Google Scholar
  37. 37.
    Whaley RC, Castaldo AM (2008) Achieving accurate and context-sensitive timing for code optimization. Software Pract Exp 38(15):1621–1642CrossRefGoogle Scholar
  38. 38.
    Whaley RC, Dongarra J (1997) Automatically Tuned Linear Algebra Software. Technical Report UT-CS-97-366, University of Tennessee, Dec 1997. http://www.netlib.org/lapack/lawns/lawn131.ps
  39. 39.
    Whaley RC, Dongarra J (1998) Automatically tuned linear algebra software. In: SuperComputing 1998: high performance networking and computing, San Antonio, TX, USA, 1998. CD-ROM Proceedings. Winner, best paper in the systems category http://www.cs.utsa.edu/~whaley/papers/atlas_sc98.ps
  40. 40.
    Whaley RC, Dongarra J (1999) Automatically Tuned Linear Algebra Software. In: Ninth SIAM conference on parallel processing for scientific computing. CD-ROM ProceedingsGoogle Scholar
  41. 41.
    Whaley RC, Petitet A Atlas homepage. http://math-atlas.sourceforge.net/
  42. 42.
    Whaley RC, Petitet A (2005) Minimizing development and maintenance costs in supporting persistently optimized BLAS. Software Pract Exp 35(2):101–121. http://www.cs.utsa.edu/~whaley/papers/spercw04.ps
  43. 43.
    Whaley RC, Petitet A, Dongarra JJ (2001) Automated empirical optimization of software and the ATLAS project. Parallel Comput 27(1–2):3–35MATHCrossRefGoogle Scholar
  44. 44.
    Whaley RC, Whalley DB (2005) Tuning high performance kernels through empirical compilation. In: The 2005 international conference on parallel processing, pp 89–98, Oslo, Norway, June 2005Google Scholar
  45. 45.
    Yi Q, Qasem A (2008) Exploring the Optimization Space of Dense Linear Algebra Kernels. In: ACM SIGPLAN symposium on library-centric software design, Aug 2008Google Scholar
  46. 46.
    Yi Q, Whaley RC (2007) Automated transformation for performance-criticial kernels. In: ACM SIGPLAN symposium on library-centric software design, Montreal, Canada, Oct 2007Google Scholar
  47. 47.
    Yotov K, Li X, Ren G, Garzaran M, Padua D, Pingali K, Stodghill P (2005) A comparison of empirical and model-driven optimization. Proceedings of the IEEE, 93(2). special issue on “Program Generation, Optimization, and Adaptation”Google Scholar

Copyright information

© Springer New York 2011

Authors and Affiliations

  1. 1.Department of Computer ScienceUniv of TXSan AntonioUSA

Personalised recommendations