Abstract
Some Linear Algebra Libraries use Level-2 routines during the factorization part of any Level-3 block factorization algorithm. We discuss four Level-3 routines called DPOTF3, a new type of BLAS, for the factorization part of a block Cholesky factorization algorithm for use by LAPACK routine DPOTRF or for BPF (Blocked Packed Format) Cholesky factorization. The four routines DPOTF3 are Fortran routines. Our main result is that performance of routines DPOTF3 is still increasing when the performance of Level-2 routine DPOTF2 of LAPACK starts to decrease. This means that the performance of DGEMM, DSYRK, and DTRSM will increase due to their use of larger block sizes and also to making less passes over the matrix elements. We present corroborating performance results for DPOTF3 versus DPOTF2 on a variety of common platforms. The four DPOTF3 routines are based on simple register blocking; different platforms have different numbers of registers and so our four routines have different register blockings. Blocked Packed Format (BPF) is discussed. LAPACK routines for _POTRF and _PPTRF using BPF instead of full and packed format are shown to be trivial modifications of LAPACK _POTRF source codes. Upper BPF is shown to be identical to square block packed format. Performance results for DBPTRF and DPOTRF for large n show that routines DPOTF3 does increase performance for large n.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Andersen, B.S., Gustavson, F.G., Waśniewski, J.: A Recursive Formulation of Cholesky Factorization of a Matrix in Packed Storage. ACM TOMS 27(2), 214–244 (2001)
Andersen, B.S., Gunnels, J.A., Gustavson, F.G., Reid, J.K., Waśniewski, J.: A Fully Portable High Performance Minimal Storage Hybrid Cholesky Algorithm. ACM TOMS 31(2), 201–227 (2005)
Anderson, E., et al.: LAPACK Users’ Guide Release 3.0. SIAM, Philadelphia (1999)
D’Azevedo, E., Dongarra, J.J.: Packed storage extension of ScaLAPACK. ORNL Report 6190, Oak Ridge National Laboratory, 13 pages (May 1998)
Dongarra, J.J., Du Croz, J., Hammarling, S., Duff, I.: Set of Level 3 Basic Linear Algebra Subprograms. TOMS 16(1), 1–17 (1990)
Elmroth, E., Gustavson, F.G., Jonsson, I., Kågström, B.: Recursive Blocked Algorithms and Hybrid Data Structures for Dense Matrix Library Software. SIAM Review 46(1), 3–45 (2004)
Gunnels, J.A., Gustavson, F.G., Pingali, K.K., Yotov, K.: Is Cache-Oblivious DGEMM Viable? In: Kågström, B., Elmroth, E., Dongarra, J., Waśniewski, J. (eds.) PARA 2006. LNCS, vol. 4699, pp. 919–928. Springer, Heidelberg (2007)
Gustavson, F.G.: Recursion Leads to Automatic Variable Blocking for Dense Linear-Algebra Algorithms. IBM J. R. & D 41(6), 737–755 (1997)
Gustavson, F.G., Jonsson, I.: Minimal Storage High Performance Cholesky via Blocking and Recursion. IBM J. R. & D 44(6), 823–849 (2000)
Gustavson, F.G.: New Generalized Data Structures for Matrices Lead to a Variety of High-Performance Algorithms. In: Boisvert, R.F., Tang, P.T.P. (eds.) Proceedings of the IFIP WG 2.5 Working Group on The Architecture of Scientific Software, Ottawa, Canada, October 2-4, pp. 211–234. Kluwer Academic Pub. (2000)
Gustavson, F.G.: High Performance Linear Algebra Algorithms using New Generalized Data Structures for Matrices. IBM J. R. & D 47(1), 31–55 (2003)
Gustavson, F.G., Gunnels, J., Sexton, J.: Minimal Data Copy For Dense Linear Algebra Factorization. In: Kågström, B., Elmroth, E., Dongarra, J., Waśniewski, J. (eds.) PARA 2006. LNCS, vol. 4699, pp. 540–549. Springer, Heidelberg (2007)
Gustavson, F.G., Reid, J.K., Waśniewski, J.: Algorithm 865: Fortran 95 Subroutines for Cholesky Factorization in Blocked Hybrid Format. ACM TOMS 33(1), 5 pages (2007)
Gustavson, F.G.: Cache Blocking. In: Jónasson, K. (ed.) PARA 2010, Part I. LNCS, vol. 7133, pp. 22–32. Springer, Heidelberg (2012)
Gustavson, F.G., Karlsson, L., Kågström, B.: Parallel and Cache-Efficient In-Place Matrix Storage Format Conversion. ACM TOMS, 34 pages (to appear, 2012)
Herrero, J.R., Navarro, J.J.: Compiler-Optimized Kernels: An Efficient Alternative to Hand-Coded Inner Kernels. In: Gavrilova, M.L., Gervasi, O., Kumar, V., Tan, C.J.K., Taniar, D., Laganá, A., Mun, Y., Choo, H. (eds.) ICCSA 2006. LNCS, vol. 3984, pp. 762–771. Springer, Heidelberg (2006)
Herrero, J.R.: New Data Structures for Matrices and Specialized Inner Kernels: Low Overhead for High Performance. In: Wyrzykowski, R., Dongarra, J., Karczewski, K., Wasniewski, J. (eds.) PPAM 2007. LNCS, vol. 4967, pp. 659–667. Springer, Heidelberg (2008)
Knuth, D.: The Art of Computer Programming, 3rd edn., vol. 1&2. Addison-Wesley
Kurzak, J., Buttari, A., Dongarra, J.: Solving systems of Linear Equations on the Cell Processor using Cholesky Factorization. IEEE Trans. Parallel Distrib. Syst. 19(9), 1175–1186 (2008)
Whaley, C.: Empirically tuning LAPACK’s blocking factor for increased performance. In: Proc. of the Conf. on Computer Aspects of Numerical Algs., 8 pages (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Gustavson, F.G., Waśniewski, J., Herrero, J.R. (2012). New Level-3 BLAS Kernels for Cholesky Factorization. In: Wyrzykowski, R., Dongarra, J., Karczewski, K., Waśniewski, J. (eds) Parallel Processing and Applied Mathematics. PPAM 2011. Lecture Notes in Computer Science, vol 7203. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-31464-3_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-31464-3_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-31463-6
Online ISBN: 978-3-642-31464-3
eBook Packages: Computer ScienceComputer Science (R0)