Registered users of NNIN/C have access to the Harvard University Faculty of Arts and Sciences (FAS) main research computing tool, the “odyssey” cluster with (as of this writing, 10/10/2012) more than 17,000 processing cores. In addition, NNIN/C at Harvard University is currently in the process of adding approximately 800 nodes to odyssey which will be available for priority use by NNIN/C members. These nodes are estimated to be installed in late November 2012.
Codes on odyssey
Software tools are built and installed as “modules” (see the modules http://www.cns.fas.harvard.edu/nnin/modules.php section of these pages) in odyssey. All of the (several hundred) odyssey modules are available to all users, and the complete list can be found here: http://rc.fas.harvard.edu/module_list/. Among these many were installed specifically for use by NNIN/C users (though now, of course, they are available to anyone who uses odyssey). Some of the principle nanoscience codes which were built for NNIN/C are:
SETE – odyssey module hpc/sete, (Single Electron Tunneling Elements) calculates electronic structure in density functional theory (DFT) of two dimensional electron gas (2DEG) based heterostructures such as quantum dots and wires at varying levels of approximation. Allows for incorporation of disorder and/or magnetic field. Calculates potential and density contours at the 2DEG level as well as eigenvalues, wave functions and tunneling coefficients for quantum dot eigenstates. Beta-versions are available for calculating electronic structure of nanowires, combination DFT and configuration interaction calculations for two electron double dots and GaAs-AlGaAs etched structures. [M. Stopa, Phys. Rev. B 54, 13767 (1996)]. This code was written and is maintained by NNIN Computation Coordinator, Mike Stopa (email@example.com). Users can contact Dr. Stopa directly for information on usage and modifications for new systems.
LAMMPS – odyssey module hpc/lamps, (Large-scale Atomic/Molecular Massively Parallel Simulator) is an open source molecular dynamics simulation code from Sandia National Laboratories. A brief description of lamps from the Sandia webpage http://lammps.sandia.gov/ is as follows:
LAMMPS is a classical molecular dynamics code that models an ensemble of particles in a liquid, solid, or gaseous state. It can model atomic, polymeric, biological, metallic, granular, and coarse-grained systems using a variety of force fields and boundary conditions.
For examples of LAMMPS simulations, see the Publications page of the LAMMPS WWW Site.
LAMMPS runs efficiently on single-processor desktop or laptop machines, but is designed for parallel computers. It will run on any parallel machine that compiles C++ and supports the MPI message-passing library. This includes distributed- or shared-memory parallel machines and Beowulf-style clusters.
LAMMPS can model systems with only a few particles up to millions or billions. See Section_perf for information on LAMMPS performance and scalability, or the Benchmarks section of the LAMMPS WWW Site.
LAMMPS is a freely-available open-source code, distributed under the terms of the GNU Public License, which means you can use or modify the code however you wish. See this section for a brief discussion of the open-source philosophy.
LAMMPS is designed to be easy to modify or extend with new capabilities, such as new force fields, atom types, boundary conditions, or diagnostics. See Section_modify for more details.
The current version of LAMMPS is written in C++. Earlier versions were written in F77 and F90. See Section_history for more information on different versions. All versions can be downloaded from the LAMMPS WWW Site.
LAMMPS was originally developed under a US Department of Energy CRADA (Cooperative Research and Development Agreement) between two DOE labs and 3 companies. It is distributed by Sandia National Labs. See this section for more information on LAMMPS funding and individuals who have contributed to LAMMPS.
In the most general sense, LAMMPS integrates Newton's equations of motion for collections of atoms, molecules, or macroscopic particles that interact via short- or long-range forces with a variety of initial and/or boundary conditions. For computational efficiency LAMMPS uses neighbor lists to keep track of nearby particles. The lists are optimized for systems with particles that are repulsive at short distances, so that the local density of particles never becomes too large. On parallel machines, LAMMPS uses spatial-decomposition techniques to partition the simulation domain into small 3d sub-domains, one of which is assigned to each processor. Processors communicate and store "ghost" atom information for atoms that border their sub-domain. LAMMPS is most efficient (in a parallel sense) for systems whose particles fill a 3d rectangular box with roughly uniform density. Papers with technical details of the algorithms used in LAMMPS are listed in this section.
Octopus – odyssey module hpc/octopus-3.2.0, is a density functional and time-dependent density functional code developed at the University of Basque. Octopus treats nuclei as point particles and uses pseudopotentials. Octopus solves the eigenvalue problem in real space. Details on octopus can be found at the site: http://www.tddft.org/programs/octopus/wiki/index.php/Main_Page.
OOMFF – odyssey module hpc/oommf, is a micromagnet simulation code developed at NIST. According to math.nist.gov/oommf:
OOMMF is a project in the Applied and Computational Mathematics Division (ACMD) of ITL/NIST, in close cooperation with µMAG, aimed at developing portable, extensible public domain programs and tools for micromagnetics. The end product will be a fully functional micromagnetic code, with a well documented, flexible programmer's interface to allow developers outside the OOMMF project to swap their own code in and out as desired. The guts of the code are being written in C++ with a Tcl/Tk (and in the future, possibly OpenGL) interface. Target systems include a wide range of Unix platforms and Windows (9x and later). The open source scripting language Tcl/Tk is required to run OOMMF.
IMOD – odyssey module hpc/imod-4.5.7, is an image reconstruction tool for 3D electron microscopy developed at the University of Colorado at Boulder. According to the imod home page (http://bio3d.colorado.edu/imod/):
IMOD is a set of image processing, modeling and display programs used for tomographic reconstruction and for 3D reconstruction of EM serial sections and optical sections. The package contains tools for assembling and aligning data within multiple types and sizes of image stacks, viewing 3-D data from any orientation, and modeling and display of the image files. IMOD was developed primarily by David Mastronarde, Rick Gaudette, Sue Held, Jim Kremer, and Quanren Xiong at the Boulder Laboratory for 3-D Electron Microscopy of Cells.
Socorro – in preparation on odyssey, from the Socorro homepage:
Socorro is a modular, object oriented code for performing self-consistent electronic-structure calculations utilizing the Kohn-Sham formulation of density-functional theory. Calculations are performed using a plane wave basis and either norm-conserving pseudopotentials or projector augmented wave functions. Several exchange-correlation functionals are available for use including the local-density approximation (Perdew-Zunger or Perdew-Wang parameterizations of the Ceperley-Alder QMC correlation results) and the generalized-gradient approximation (PW91, PBE, and BLYP). Both Fourier-space and real-space projectors have been implemented, and a variety of methods are available for relaxing atom positions, optimizing cell parameters, and performing molecular dynamics calculations.
Other programs on odyssey of possible interest to nanoscientists: Turbomole (ab initio code for molecules), Charmm (molecular dynamics code), Gaussian (ab inito code, Gaussian basis functions). Also various utilities such as Fast Fourier Transform. See the odyssey modules page for a complete list.
A number of codes which have been built and used on NNIN/C clusters in the past are currently unavailable. Users interested in these tools can contact NNIN/C to determine the feasibility of resurrecting these codes.
XDTrimSP – Monte-Carlo simulation of ion implantation in solids.
HARES (High performance fortran Adaptive grid Real space Electronic Structure) calculates electronic structure of crystals and small molecules using a real space, adaptive grid. Its real-space character makes HARES highly parallelizable and applicable to nanoscale systems with varying boundary conditions [Waghmere et al., cond-mat/0006183].
EDIP (Environment Dependent Interatomic Potential) is an efficient and realistic model for interatomic forces in covalent solids and liquids which incorporates recent theoretical advances in understanding the environment dependence of (sigma) chemical bonding in condensed phases. Recently, EDIP has been extended by N.A. Marks to carbon by incorporating the effects of pi-bonding empirically [N. A. Marks, Phys. Rev. B 63 035401 (2001), M. Bazant et al., Phys. Rev. B 56, 8542 (1997)].
ANEBA (Adaptive Nudged Elastic Band Approach) locates the saddle point in the potential energy surface between an initial and a final state in a physical transition process such as a chemical reaction or diffusion process.