Abstract
Various software automatic tuning methods have been proposed to search for the optimum parameter setting from among a combination of performance parameters. We have been studying a discrete spline (d-Spline)-based incremental performance parameter estimation (IPPE) method that does not require the approximation function to have differential continuity. In this method, a d-Spline generated from the minimum sample point is used to estimate the optimum value of the performance parameter. In prior methods, one measurement result was used to conduct sample point estimation; however, perturbations arising from the computing environment can affect estimates made in this manner. Such perturbations include disturbances introduced by the computing environment and OS jitters. In this study, we propose a method that considers execution time perturbation in performance parameter estimation by allowing for re-measurement under certain conditions by using an actual IPPE measurement. This lowers the inclusion of execution time perturbation in d-Spline approximation, thus enhancing the reliability of software automatic tuning.
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
References
Clint Whaley, R., Petitet, A., Dongarra, J.J.: Automated empirical optimization of software and ATLAS project. Parallel Comput. 27, 3–35 (2001)
Suda, R.: A Bayesian method of online automatic tuning. In: Naono, K., Teranishi, K., Cavazos, J., Suda, R. (eds.) Software Automatic Tuning, pp. 275–293. Springer, New York (2011). https://doi.org/10.1007/978-1-4419-6935-4_16
Chen, J., Che, R., Fujii, A., Suda, R., Wang, W.: Timing performance surrogates in auto-tuning for qualitative and quantitative factors. In: SIAM Conference on Parallel Processing and Scientific Computing, PP 2014 (2014)
Katagiri, T., Ito, S., Ohshima, S.: Early experiences for adaptation of auto-tuning by ppOpen-AT to an explicit method. In: Proceedings of MCSoC 2013, pp. 153–158 (2013)
Mochizuki, M., Fujii, A., Tanaka, T.: Fast multidimensional performance parameter estimation with multiple one-dimensional d-Spline parameter search. In: iWAPT (2017)
Katagiri, T., Ohshima, S., Matsumoto, M.: Auto-tuning of computation kernels from an FDM code with ppOpen-AT. In: Proceedings of MCSoC 2014, pp. 91–98 (2014)
ppOpen-HPC Project. http://ppopenhpc.cc.u-tokyo.ac.jp/ppopenhpc. Accessed 15 Feb 2017
Murata, R., Irie, J., Fujii, A., Tanaka, T., Katagiri, T.: Enhancement of incremental performance parameter estimation on ppOpen-AT. In: Proceedings of MCSoC 2015, pp. 203–210 (2015)
Tanaka, T., Katagiri, T., Yuba, T.: d-Spline based incremental parameter estimation in automatic performance tuning. In: Kågström, B., Elmroth, E., Dongarra, J., Waśniewski, J. (eds.) PARA 2006. LNCS, vol. 4699, pp. 986–995. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-75755-9_116
Tanaka, T., Otsuka, R., Fujii, A., Katagiri, T., Imamura, T.: Implementation of d-Spline-based incremental performance parameter estimation method with ppOpen-AT. Sci. Program. 22, 299–307 (2014)
Vanek, P., Mandel, J., Brezina, M.: Algebraic multigrid by smoothed aggregation for second and fourth order elliptic problems. Technical report UCD-CCM-036 (1995)
Acknowledgments
This study was partially supported by JSPS KAKENHI Grant Number JP 16H02823,15H02708, and JSPS, Open Partnership Joint Research Projects/Seminars, “Deepening Performance Models for Automatic Tuning with International Collaboration.”
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG, part of Springer Nature
About this paper
Cite this paper
Fan, G., Mochizuki, M., Fujii, A., Tanaka, T., Katagiri, T. (2018). D-Spline Performance Tuning Method Flexibly Responsive to Execution Time Perturbation. In: Wyrzykowski, R., Dongarra, J., Deelman, E., Karczewski, K. (eds) Parallel Processing and Applied Mathematics. PPAM 2017. Lecture Notes in Computer Science(), vol 10777. Springer, Cham. https://doi.org/10.1007/978-3-319-78024-5_34
Download citation
DOI: https://doi.org/10.1007/978-3-319-78024-5_34
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-78023-8
Online ISBN: 978-3-319-78024-5
eBook Packages: Computer ScienceComputer Science (R0)