Scalable Reactive Molecular Dynamics Simulations for Computational Synthesis
Creators
Abstract
Reactive molecular dynamics (MD) simulation is a powerful research tool for describing chemical reactions. We eliminate the speed-limiting charge iteration in MD with a novel extended-Lagrangian scheme. The extended-Lagrangian reactive MD (XRMD) code drastically improves energy conservation while substantially reducing time-to-solution. Furthermore, we introduce a new polarizable charge equilibration (PQEq) model to accurately predict atomic charges and polarization. The XRMD code based on hybrid message passing+multithreading achieves a weak-scaling parallel efficiency of 0.977 on 786 432 IBM Blue Gene/Q cores for a 67.6 billion-atom system. The performance is portable to the second-generation Intel Xeon Phi, Knights Landing. Blue Gene/Q simulations for the computational synthesis of materials via novel exfoliation mechanisms for synthesizing atomically thin transition metal dichalcogenide layers will dominate nanomaterials science in this century.
Additional Information
© 2018 IEEE. Published by the IEEE Computer Society. Date of publication 11 January 2018; date of current version 15 August 2019. This work was supported as part of the Computational Materials Sciences Program funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, under Award DE-SC00014607. The authors would like to thank P. E. Small, P. Rajak, S. Tiwari, G. Zhou, and S. Naserifar for their help. This research used resources of the Argonne Leadership Computing Facility (ALCF), a U.S. Department of Energy Office of Science User Facility operated under contract DE-AC02-06CH-11357. The work of Y. Li was supported by the Margaret Butler Postdoctoral Fellowship at ALCF.Attached Files
Accepted Version - 1261-abs-IEEE-Scalable_Reactive_Molecular_Dynamics.pdf
Files
1261-abs-IEEE-Scalable_Reactive_Molecular_Dynamics.pdf
Files
(6.6 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:1a97c14303b91f96a4d83d5391e61b90
|
6.6 MB | Preview Download |
Additional details
Identifiers
- Eprint ID
- 85685
- DOI
- 10.1109/MCSE.2018.110150043
- Resolver ID
- CaltechAUTHORS:20180406-113403066
Related works
- Describes
- 10.1109/MCSE.2018.110150043 (DOI)
Funding
- Department of Energy (DOE)
- DE-SC00014607
- Department of Energy (DOE)
- DE-AC02-06CH-11357
- Argonne Leadership Computing Facility
Dates
- Created
-
2018-04-06Created from EPrint's datestamp field
- Updated
-
2023-06-08Created from EPrint's last_modified field