A massively parallel algorithm for grand canonical Monte Carlo computer simulation with the short-ranged Lennard-Jones potential
- Sandia National Labs., Albuquerque, NM (United States)
- NCUBE Corp., Sandia Park, NM (United States)
We present a new massively parallel decomposition for grand canonical Monte Carlo computer simulation (GCMC) suitable for short ranged fluids. Our spatial algorithm relies on the fact that for short-ranged fluids, molecules separated by a greater distance than the reach of the potential act independently, thus different processors can work concurrently in regions of the same system which are sufficiently far apart. Several parallelization issues unique to GCMC are addressed such as the handling of the three different types of Monte Carlo move used in GCMC: the displacement of a molecule, the creation of a molecule, and the destruction of a molecule. The decomposition is shown to scale with system size, making it especially useful for systems where the physical problem dictates the system size, for example, fluid behavior in mesopores.
- Research Organization:
- Sandia National Labs., Albuquerque, NM (United States)
- Sponsoring Organization:
- USDOE, Washington, DC (United States)
- DOE Contract Number:
- AC04-94AL85000
- OSTI ID:
- 10144429
- Report Number(s):
- SAND--94-0989C; CONF-941118--1; ON: DE94010396; BR: GB0103012
- Country of Publication:
- United States
- Language:
- English
Similar Records
Monte Carlo simulation of supercritical solutions in the grand canonical ensemble
A Grand Canonical Monte Carlo simulation program for computing ion distributions around biomolecules in hard sphere solvents