Path-accelerated stochastic molecular dynamics: Parallel-in-time integration using path integrals
© 2019 Author(s). Massively parallel computer architectures create new opportunities for the performance of long-time scale molecular dynamics (MD) simulations. Here, we introduce the path-accelerated molecular dynamics method that takes advantage of distributed computing to reduce the wall-clock ti...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIP Publishing
2021
|
| Online Access: | https://hdl.handle.net/1721.1/132218 |
| _version_ | 1811085028558372864 |
|---|---|
| author | Rosa-Raíces, Jorge L Zhang, Bin Miller, Thomas F |
| author_facet | Rosa-Raíces, Jorge L Zhang, Bin Miller, Thomas F |
| author_sort | Rosa-Raíces, Jorge L |
| collection | MIT |
| description | © 2019 Author(s). Massively parallel computer architectures create new opportunities for the performance of long-time scale molecular dynamics (MD) simulations. Here, we introduce the path-accelerated molecular dynamics method that takes advantage of distributed computing to reduce the wall-clock time of MD simulation via parallelization with respect to stochastic MD time steps. The marginal distribution for the time evolution of a system is expressed in terms of a path integral, enabling the use of path sampling techniques to numerically integrate MD trajectories. By parallelizing the evaluation of the path action with respect to time and by initializing the path configurations from a nonequilibrium distribution, the algorithm enables significant speedups in terms of the length of MD trajectories that can be integrated in a given amount of wall-clock time. The method is demonstrated for Brownian dynamics, although it is generalizable to other stochastic equations of motion including open systems. We apply the method to two simple systems, a harmonic oscillator and a Lennard-Jones liquid, and we show that in comparison to the conventional Euler integration scheme for Brownian dynamics, the new method can reduce the wall-clock time for integrating trajectories of a given length by more than three orders of magnitude in the former system and more than two in the latter. This new method for parallelizing MD in the dimension of time can be trivially combined with algorithms for parallelizing the MD force evaluation to achieve further speedup. |
| first_indexed | 2024-09-23T13:01:48Z |
| format | Article |
| id | mit-1721.1/132218 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T13:01:48Z |
| publishDate | 2021 |
| publisher | AIP Publishing |
| record_format | dspace |
| spelling | mit-1721.1/1322182021-09-21T03:32:22Z Path-accelerated stochastic molecular dynamics: Parallel-in-time integration using path integrals Rosa-Raíces, Jorge L Zhang, Bin Miller, Thomas F © 2019 Author(s). Massively parallel computer architectures create new opportunities for the performance of long-time scale molecular dynamics (MD) simulations. Here, we introduce the path-accelerated molecular dynamics method that takes advantage of distributed computing to reduce the wall-clock time of MD simulation via parallelization with respect to stochastic MD time steps. The marginal distribution for the time evolution of a system is expressed in terms of a path integral, enabling the use of path sampling techniques to numerically integrate MD trajectories. By parallelizing the evaluation of the path action with respect to time and by initializing the path configurations from a nonequilibrium distribution, the algorithm enables significant speedups in terms of the length of MD trajectories that can be integrated in a given amount of wall-clock time. The method is demonstrated for Brownian dynamics, although it is generalizable to other stochastic equations of motion including open systems. We apply the method to two simple systems, a harmonic oscillator and a Lennard-Jones liquid, and we show that in comparison to the conventional Euler integration scheme for Brownian dynamics, the new method can reduce the wall-clock time for integrating trajectories of a given length by more than three orders of magnitude in the former system and more than two in the latter. This new method for parallelizing MD in the dimension of time can be trivially combined with algorithms for parallelizing the MD force evaluation to achieve further speedup. 2021-09-20T18:21:23Z 2021-09-20T18:21:23Z 2020-09-23T12:53:50Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/132218 en 10.1063/1.5125455 The Journal of Chemical Physics Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf AIP Publishing arXiv |
| spellingShingle | Rosa-Raíces, Jorge L Zhang, Bin Miller, Thomas F Path-accelerated stochastic molecular dynamics: Parallel-in-time integration using path integrals |
| title | Path-accelerated stochastic molecular dynamics: Parallel-in-time integration using path integrals |
| title_full | Path-accelerated stochastic molecular dynamics: Parallel-in-time integration using path integrals |
| title_fullStr | Path-accelerated stochastic molecular dynamics: Parallel-in-time integration using path integrals |
| title_full_unstemmed | Path-accelerated stochastic molecular dynamics: Parallel-in-time integration using path integrals |
| title_short | Path-accelerated stochastic molecular dynamics: Parallel-in-time integration using path integrals |
| title_sort | path accelerated stochastic molecular dynamics parallel in time integration using path integrals |
| url | https://hdl.handle.net/1721.1/132218 |
| work_keys_str_mv | AT rosaraicesjorgel pathacceleratedstochasticmoleculardynamicsparallelintimeintegrationusingpathintegrals AT zhangbin pathacceleratedstochasticmoleculardynamicsparallelintimeintegrationusingpathintegrals AT millerthomasf pathacceleratedstochasticmoleculardynamicsparallelintimeintegrationusingpathintegrals |