Computing Correlation Times for Parallel Replica Methods

Material Information

Computing Correlation Times for Parallel Replica Methods
Rufa, Dominic Antonio
Publication Date:


Parallel Replica Dynamics, and the more recent Parallel Trajectory Splicing, are promising time parallelization methods that have been shown to accelerate the dynamics of several classes of atomistic systems characterized by infrequent event transition events. However, perhaps the most notable obstacle inhibiting the extensibility of these methods to arbitrary systems is the dilemma associated with estimating the relaxation time for replica ensembles within metastable states. This report presents three methods to accurately calculate this relaxation, or correlation time (corr), for a class of systems. Furthermore, simulations are conducted on both model potentials and Ala3 in silico to illustrate the advantages and limitations of each method. ( en )
General Note:
Awarded Bachelor of Science, magna cum laude, Major: Chemistry, and Bachelor of Science, cum laude, Major: Physics on, May 8, 2018.
General Note:
College or School: College of Liberal Arts and Sciences
General Note:
Advisor: Adrian Roitberg. Advisor Department or School: Chemistry

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Dominic Antonio Rufa. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

UFDC Membership

UF Undergraduate Honors Theses


This item is only available as the following downloads:

Full Text