An Improved Molecular Dynamics Algorithm for Large Scale Simulation in Nano-engineering

Focusing on the special need of nano-engineering, an improved neighbor list algorithm for the ultra-large scale molecular dynamics (MD) simulation is proposed. The new algorithm is attempting to fully combine the advantages of both Verlet table and cell-linked list algorithms to accelerate the speed...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Tang, Yu-lan, Chen, Jia-xuan
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Focusing on the special need of nano-engineering, an improved neighbor list algorithm for the ultra-large scale molecular dynamics (MD) simulation is proposed. The new algorithm is attempting to fully combine the advantages of both Verlet table and cell-linked list algorithms to accelerate the speed of neighbor list construction and update. In order to meet the requirement of nano-engineering that often requires the frequent update of the cell-linked lists and neighbor lists, the program employs the pointer link data structure so that both reliability and accuracy of the simulation can be ensured. The performance of molecular dynamics simulation is evaluated using the proposed algorithm and compared with those using conventional Verlet table and cell-linked list algorithms. Results show that the new algorithm outperforms the conventional cell-linked list algorithm by 2~3 times for systems of 10 4 ~ 10 6 atoms per single CPU.
DOI:10.1109/ICICTA.2009.438