Abstract
In this paper, we propose a process grid free algorithm for a massively parallel dense symmetric eigensolver with a communication splitting multicasting algorithm. In this algorithm, a tradeoff exists between speed and memory space to keep the Householder vectors. As a result of a performance evaluation with the T2K Open Supercomputer (U. Tokyo) and the RX200S5, we obtain the performance with 0.86x and 0.95x speed-downs and 1/2 memory space compared to the conventional algorithm for a square process grid. We also show a new algorithm for small-sized matrices in massively parallel processing that takes an appropriately small value of p of the process grid p x q. In this case, the execution time of inverse transformation is negligible.
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
Inaba, T., Tsunekawa, N., Hirano, T., Yoshihiro, T., Kashiwagi, H., Sato, F.: Density Functional Calculation of the Electronic Structure on Insulin Hexamer. Chemical Physics Letters 434(4-6), 331–335 (2007)
Dongarra, J.J., Hammarling, S.J., Sorensen, D.C.: Block Reduction of Matrices to Condensed Forms for Eigenvalue Computations. Journal of Computational and Applied Mathematics 27, 215–227 (1989)
Bischof, C.H., Marques, M., Sun, X.: Parallel Bandreduction and Tridiagonalization. In: Proceedings of Sixth SIAM Conference on Parallel Processing for Scientific Computing, pp. 22–24 (1993)
Katagiri, T., Kanada, Y.: An Efficient Implementation of Parallel Eigenvalue Computation for Massively Parallel Processing. Parallel Computing 27(14), 1831–1845 (2001)
Katagiri, T., Kise, K., Honda, H., Yuba, T.: ABCLib_DRSSED: A Parallel Eigensolver with an Auto-tuning Facility. Parallel Computing 32(3), 231–250 (2006)
ABCLib_DRSSED home page, http://www.abc-lib.org/main1.html
Imamura, T.: How To Develop The Eigenvalue Solver Which Organizes Beyond Hundred Thousand Cores. IPSJ SIG Notes 2009-HPC-121(19) (2009) (in Japanese)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Katagiri, T., Itoh, S. (2011). A Massively Parallel Dense Symmetric Eigensolver with Communication Splitting Multicasting Algorithm. In: Palma, J.M.L.M., Daydé, M., Marques, O., Lopes, J.C. (eds) High Performance Computing for Computational Science – VECPAR 2010. VECPAR 2010. Lecture Notes in Computer Science, vol 6449. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-19328-6_15
Download citation
DOI: https://doi.org/10.1007/978-3-642-19328-6_15
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-19327-9
Online ISBN: 978-3-642-19328-6
eBook Packages: Computer ScienceComputer Science (R0)