Computation at Stanford
The Nanotechnology Computation Facility at Stanford provides expertise, collaborations, and consulting services in the following areas:
- Atomistic nanoelectronics simulations, including ab initio NEGF transport simulations.
- Atomistic simulations related to energy sciences: hydrogen storage, photovoltaic, nuclear energy, and heterogonous and homogeneous catalysis
- Expertise in both plane wave and local basis set based DFT methods and wavefunction based methods such as Configuration Interaction (CI), Coupled Cluster (CC), Many Body Perturbation (MP), and Quantum Monte Carlo (QMC).
- Ab Initio and Force Field based molecular dynamics simulation
- Atomistic simulation of biological systems
There are 64 duel processor nodes for a total of 512 cores. The CPU is Intel Quad-Core Xeon Processor E5440 (12M Cache, 2.83 GHz, 1333 MHz FSB). All 64 nodes are linked with Gigabit network and 24 nodes are linked with Infiniband. Each node has 16 GB of DDR2-667 RAM. The total storage space is ca. 11TB.
Installed application software:
It is written in a C++ object oriented framework for the conformational/sequence sampling/optimization of biophysical structures (proteins, nucleic acids). Molecular entities/biopolymers can be described at different resolution from one center per monomer (residue/nucleotide) to all atom description with solvent effects. It is free of charge, after registration with the developer.
CHAMP: performs Quantum Monte Carlo calculations. It is free but requires agreement with developer partly due to the sophisticated nature of the method.
CP2K: performs atomistic and molecular simulations of solid state, liquid, molecular and biological systems. It uses a mixed Gaussian and plane waves approach (GPW), and classical pair and many-body potential. It is freely available.
CPMD: designed for ab-initio molecular dynamics based on parallelized plane wave/pseudopotential implementation of Density Functional Theory. It is freely available.
GAMESS: one of the most popular quantum chemistry software packages and is widely used for chemistry and biology applications. It includes density functional theory, and other more sophisticated correlation methods including configuration interaction, perturbation theory, and coupled cluster theory. It is freely available.
Gaussian: one of the earliest and most popular quantum chemistry software packages.
It is commercial and Stanford, like many other institutions, has a site license. It is the software package that a large percentage of users, both theoretical and experimental, in chemistry are first introduced to.
NWChem: a comprehensive software package that includes most of the functionality typically used in chemistry, biology, and material sciences. It is designed for massively parallel computers and is highly scalable. It is freely available.
VASP: one of the standard simulation tools based on pseudopotential and plane wave basis set for solid state physics. It is licensed software and a few groups at Stanford may have license to it.
ATK/VNL: designed for electronic transport calculations based on DFT and local basis set. It is commercial.
GROMACS: high performance molecular dynamics program for bio-molecules.
Desmond: high-speed molecular dynamics simulations of biological systems.
Installed compilers and system libraries:
FORTRAN and C/C++ compilers, MKL and FTT libraries from Intel
Mpich2 compilers and libraries with socket and shared memory implementation
Installed Visualization software:
Molden and ecce: used to visualize and build structures, plot wavefunction and charge densities, visualize molecular orbitals, and visualize vibrational movements. Ecce also has extensive structure database for typical structures and structure fragments frequently used in chemistry and biology. These are free.