skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A generic, hierarchical framework for massively parallel Wang Landau sampling

Abstract

We introduce a parallel Wang Landau method based on the replica-exchange framework for Monte Carlo simulations. To demonstrate its advantages and general applicability for simulations of com- plex systems, we apply it to the self-assembly process in amphiphilic solutions and to lattice protein adsorption. Without loss of accuracy, the method gives significant speed-up on small architectures like multi-core processors, and should be beneficial for petaflop machines.

Authors:
 [1];  [2];  [3];  [1]
  1. University of Georgia, Athens, GA
  2. ORNL
  3. Swiss Federal Research Institute, Switzerland
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE Office of Science (SC)
OSTI Identifier:
1092241
DOE Contract Number:
DE-AC05-00OR22725
Resource Type:
Journal Article
Resource Relation:
Journal Name: Physical Review Letters; Journal Volume: 110; Journal Issue: 21
Country of Publication:
United States
Language:
English

Citation Formats

Vogel, Thomas, Li, Ying Wai, Wuest, Thomas, and Landau, David P. A generic, hierarchical framework for massively parallel Wang Landau sampling. United States: N. p., 2013. Web. doi:10.1103/PhysRevLett.110.210603.
Vogel, Thomas, Li, Ying Wai, Wuest, Thomas, & Landau, David P. A generic, hierarchical framework for massively parallel Wang Landau sampling. United States. doi:10.1103/PhysRevLett.110.210603.
Vogel, Thomas, Li, Ying Wai, Wuest, Thomas, and Landau, David P. Tue . "A generic, hierarchical framework for massively parallel Wang Landau sampling". United States. doi:10.1103/PhysRevLett.110.210603.
@article{osti_1092241,
title = {A generic, hierarchical framework for massively parallel Wang Landau sampling},
author = {Vogel, Thomas and Li, Ying Wai and Wuest, Thomas and Landau, David P},
abstractNote = {We introduce a parallel Wang Landau method based on the replica-exchange framework for Monte Carlo simulations. To demonstrate its advantages and general applicability for simulations of com- plex systems, we apply it to the self-assembly process in amphiphilic solutions and to lattice protein adsorption. Without loss of accuracy, the method gives significant speed-up on small architectures like multi-core processors, and should be beneficial for petaflop machines.},
doi = {10.1103/PhysRevLett.110.210603},
journal = {Physical Review Letters},
number = 21,
volume = 110,
place = {United States},
year = {Tue Jan 01 00:00:00 EST 2013},
month = {Tue Jan 01 00:00:00 EST 2013}
}
  • Wang Landau sampling is implemented on the Graphics Processing Unit (GPU) with the Compute Unified Device Architecture (CUDA). Performances on three different GPU cards, including the new generation Fermi architecture card, are compared with that on a Central Processing Unit (CPU). The parameters for massively parallel Wang Landau sampling are tuned in order to achieve fast convergence. For simulations of the water cluster systems, we obtain an average of over 50 times speedup for a given workload.
  • We investigate a generic, parallel replica-exchange framework for Monte Carlo simulations based on the Wang Landau method. To demonstrate its advantages and general applicability for massively parallel simulations of complex systems, we apply it to lattice spin models, the self-assembly process in amphiphilic solutions, and the adsorption of molecules on surfaces. While of general, current interest, the latter phenomena are challenging to study computationally because of multiple structural transitions occurring over a broad temperature range. We show how the parallel framework facilitates simulations of such processes and, without any loss of accuracy or precision, gives a significant speedup and allowsmore » for the study of much larger systems and much wider temperature ranges than possible with single-walker methods.« less
  • Cited by 23
  • Generic features associated with the adsorption of proteins on solid surfaces are reviewed within the framework of the hydrophobic-polar (HP) lattice protein model. The thermodynamic behavior and structural properties of various HP protein sequences interacting with attractive surfaces have been studied using extensive Wang-Landau sampling with different types of surfaces, each of which attracts either: all monomers, only hydrophobic (H) monomers, or only polar (P) monomers, respectively. Consequently, different types of folding behavior occur for varied surface strengths. Analysis of the combined patterns of various structural observables, e.g., the derivatives of the numbers of interaction contacts, together with the specificmore » heat, leads to the identification of fundamental categories of folding and transition hierarchies. We also inferred a connection between the transition categories and the relative surface strengths, i.e., the ratios of the surface attractive strengths to the intra-chain attraction among H monomers. We thus believe that the folding hierarchies and identification scheme are generic for different HP sequences interacting with attractive surfaces, regardless of the chain length, sequence, or surface attraction.« less