The Shift-and-invert parallel spectral transformations (SIPs), a computational approach to solve sparse eigenvalue problems, is developed for massively parallel architectures with exceptional parallel scalability and robustness. The capabilities of SIPs are demonstrated by diagonalization of density-functional based tight-binding (DFTB) Hamiltonian and overlap matrices for single-wall metallic carbon nanotubes, diamond nanowires, and bulk diamond crystals. The largest (smallest) example studied is a 128,000 (2000) atom nanotube for which ~330,000 (~5600) eigenvalues and eigenfunctions are obtained in ~190 (~5) seconds when parallelized over 266,144 (16,384) Blue Gene/Q cores. Weak scaling and strong scaling of SIPs are analyzed and the performance of SIPs is compared with other novel methods. Different matrix ordering methods are investigated to reduce the cost of the factorization step, which dominates the time-to-solution at the strong scaling limit. As a result, a parallel implementation of assembling the density matrix from the distributed eigenvectors is demonstrated.
Zhang, Hong, Zapol, Peter, Dixon, David A., Wagner, Albert F., & Keceli, Murat (2015). Shift-and-invert parallel spectral transformation eigensolver: Massively parallel performance for density-functional based tight-binding. Journal of Computational Chemistry, 37(4). https://doi.org/10.1002/jcc.24254
Zhang, Hong, Zapol, Peter, Dixon, David A., et al., "Shift-and-invert parallel spectral transformation eigensolver: Massively parallel performance for density-functional based tight-binding," Journal of Computational Chemistry 37, no. 4 (2015), https://doi.org/10.1002/jcc.24254
@article{osti_1248169,
author = {Zhang, Hong and Zapol, Peter and Dixon, David A. and Wagner, Albert F. and Keceli, Murat},
title = {Shift-and-invert parallel spectral transformation eigensolver: Massively parallel performance for density-functional based tight-binding},
annote = {The Shift-and-invert parallel spectral transformations (SIPs), a computational approach to solve sparse eigenvalue problems, is developed for massively parallel architectures with exceptional parallel scalability and robustness. The capabilities of SIPs are demonstrated by diagonalization of density-functional based tight-binding (DFTB) Hamiltonian and overlap matrices for single-wall metallic carbon nanotubes, diamond nanowires, and bulk diamond crystals. The largest (smallest) example studied is a 128,000 (2000) atom nanotube for which ~330,000 (~5600) eigenvalues and eigenfunctions are obtained in ~190 (~5) seconds when parallelized over 266,144 (16,384) Blue Gene/Q cores. Weak scaling and strong scaling of SIPs are analyzed and the performance of SIPs is compared with other novel methods. Different matrix ordering methods are investigated to reduce the cost of the factorization step, which dominates the time-to-solution at the strong scaling limit. As a result, a parallel implementation of assembling the density matrix from the distributed eigenvectors is demonstrated.},
doi = {10.1002/jcc.24254},
url = {https://www.osti.gov/biblio/1248169},
journal = {Journal of Computational Chemistry},
issn = {ISSN 0192-8651},
number = {4},
volume = {37},
place = {United States},
publisher = {Wiley},
year = {2015},
month = {11}}
Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 372, Issue 2011https://doi.org/10.1098/rsta.2012.0483