Abstract
Multicore systems are becoming ubiquituous in scientific computing. As performance libraries are adapted to such systems, the difficulty to extract the best performance out of them is quite high. Indeed, performance libraries such as Intel’s MKL, while performing very well on unicore architectures, see their behaviour degrade when used on multicore systems. Moreover, even multicore systems show wide differences among each other (presence of shared caches, memory bandwidth, etc.) We propose a systematic method to improve the parallel execution of matrix multiplication, through the study of the behavior of unicore DGEMM kernels in MKL, as well as various other criteria. We show that our fine-tuning can out-perform Intel’s parallel DGEMM of MKL, with performance gains sometimes up to a factor of two.
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
Asanovic, K., Bodik, R., Catanzaro, B.C., Gebis, J.J., Husbands, P., Keutzer, K., Patterson, D.A., Plishker, W.L., Shalf, J., Williams, S.W., Yelick, K.A.: The landscape of parallel computing research: A view from berkeley. Technical report, EECS Department, Univ. of California, Berkeley (December 2006)
Cannon, L.E.: A cellular computer to implement the kalman filter algorithm. Ph.D thesis (1969)
Chan, E., Quintana-Orti, E.S., Quintana-Orti, G., van de Geijn, R.: Supermatrix out-of-order scheduling of matrix operations for smp and multi-core architectures. In: SPAA 2007: Proceedings of the nineteenth annual ACM symposium on Parallel algorithms and architectures, pp. 116–125. ACM, New York (2007)
Fox, G.C., Furmanski, W., Walker, D.W.: Optimal matrix algorithms on homogeneous hypercubes. In: Proceedings of the 3rd conference on Hypercube concurrent computers and applications. ACM, New York (1988)
Goto, K., van de Geijn, R.: High performance implementation of the level-3. Transactions on Mathematical Software 35(1) (2008)
Goto, K., van de Geijn, R.A.: Anatomy of a high-performance matrix multiplication. Transactions on Mathematical Software 34(3) (2008)
Krishnan, M., Nieplocha, J.: Srumma: A matrix multiplication algorithm suitable for clusters and scalable shared memory systems. In: IPDPS (2004)
Marc Pérache, H.J., Namyst, R.: Mpc: a unified parallel runtime for clusters of numa machines. In: Luque, E., Margalef, T., Benítez, D. (eds.) Euro-Par 2008. LNCS, vol. 5168, pp. 78–88. Springer, Heidelberg (2008)
Matthias Christen, O.S., Burkhart, H.: Graphical processing units as co-processors for hardware-oriented numerical solvers. In: Workshop PARS 2007 (2006)
Whaley, R.C., Petitet, A., Dongarra, J.J.: Automated empirical optimizations of software and the ATLAS project. In: Parallel Computing (2001)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zuckerman, S., Pérache, M., Jalby, W. (2008). Fine Tuning Matrix Multiplications on Multicore. In: Sadayappan, P., Parashar, M., Badrinath, R., Prasanna, V.K. (eds) High Performance Computing - HiPC 2008. HiPC 2008. Lecture Notes in Computer Science, vol 5374. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-89894-8_7
Download citation
DOI: https://doi.org/10.1007/978-3-540-89894-8_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-89893-1
Online ISBN: 978-3-540-89894-8
eBook Packages: Computer ScienceComputer Science (R0)