Premium
Characterization of a hybrid Monte Carlo search algorithm for structure determination
Author(s) -
Shankland Kenneth,
David William I. F.,
Didlick Gareth,
Markvardsen Anders J.
Publication year - 2005
Publication title -
journal of applied crystallography
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.429
H-Index - 162
ISSN - 1600-5767
DOI - 10.1107/s0021889804028894
Subject(s) - monte carlo method , algorithm , frame (networking) , computer science , function (biology) , characterization (materials science) , hybrid monte carlo , diffraction , monte carlo algorithm , statistical physics , markov chain monte carlo , physics , mathematics , optics , statistics , telecommunications , evolutionary biology , biology
A hybrid Monte Carlo (HMC) search algorithm has recently been shown to be a promising method for structure determination from powder diffraction data [Johnston, David, Markvardsen & Shankland (2002). Acta Cryst. A 58 , 441–447]. Here, the performance of the algorithm on a number of different crystal structures is investigated as a function of its control parameters. This detailed analysis required the use of a system for distributed computing in order to keep the calculation times within a reasonable time frame. The results obtained confirm previous findings and a detailed discussion of the effect of the control parameters on the efficiency of the HMC method is provided. The results suggest a method for setting these parameters automatically, which is an essential step if HMC is to find routine use in the determination of crystal structures.