MATERIALS THEORY GROUP

High-performance computing facilities

MTG has assess to the high-performance supercomputer at the Korea Institute of Science & Technology Information Supercomputing Center (KISTI-KSC). Following the service retirement of KISTI's 5th-generation supercomputer Nurion (누리온), computing resources are powered by KISTI's 6th-generation national supercomputer Hangang (한강) and the GPU-accelerated Neuron (뉴론) cluster.

Hangang is a 614 PFlops theoretical peak performance supercomputer built by HPE with 205 PB of storage. The system is configured with 8,496 NVIDIA GPUs—comprising 8,336 NVIDIA GH200 Grace Hopper superchips and 160 NVIDIA H200 SXM GPUs across 20 high-performance GPU-Fat nodes and 9,936 CPUs, featuring 800 CPU nodes powered by dual 5th Gen AMD EPYC "Turin" 9745 processors (204,800 total CPU cores at 2.4 GHz) in addition to ARM Grace CPUs.

Neuron serves as KISTI's specialized heterogeneous computing cluster, providing multi-tier GPU partitions (featuring NVIDIA V100, A100, H100, H200, and GH200 Grace Hopper nodes) as well as dedicated CPU and high-memory partitions to support deep learning, AI foundation model training, and data-intensive calculations.

In-house computing cluster, KOHN

Our group owns an in-house InfiniBand connected Intel cluster of about 80 nodes, making up a total of approximately close to 2,000 CPU cores. This includes 2 GPU nodes (4 units of GeForce RTX™ 3090 Ti) with 32 CPU and 43,008 CUDA cores, and 2 GPU nodes of the recent RTX PRO 6000 Blackwell Max-Q Workstation Edition.

In addition, our group also has assess to high performance supercomputing facilities in Australia, through a long-term, active collaboration with the Condensed Matter Theory group (led by Professor Catherine Stampfl FAA FRSN) in the School of Physics, University of Sydney, Australia: the NCI National Facility in Canberra, and the Pawsey Supercomputing Centre in Perth.

Density-functional theory codes

Vienna Ab initio Simulation Package (VASP) code
Fritz Haber Institute - Ab Initio Molecular Simulations (FHI-aims) code
Quantum opEn-Source Package for Research in Electronic Structure, Simulation, and Optimization: Quantum-ESPRESSO package
All-electron full-potential linearised augmented-plane wave (FP-LAPW) "Electrons in k-space" Elk code
Density functional for molecules and three-dimensional periodic solids (DMol3) code
Octopus code for time-dependent density-functional theory (TD-DFT) calculations
Cambridge Serial Total Energy Package (CASTEP) code

Molecular dynamics codes

Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) [Tutorials].
GROMACS: A free and open-source software suite for high-performance molecular dynamics and output analysis [Tutorials].
General Utility Lattice Program (GULP).
i-PI (a Python force engine) was initially developed for Path Integral Molecular Dynamics (PIMD) simulations, and contains probably the most comprehensive array of PIMD techniques. However, it has evolved over time to become a general-purpose code with methods ranging from phonon calculators to replica exchange molecular dynamics. It can be used to run simulations using empirical forcefields or machine learning interatomic potentials with tens of thousands of atoms with only a small overhead [Documentation].
Open-source software for parallel molecular simulations (PIMD).
TRajectory Analyzer and VISualizer (TRAVIS): Tool for analyzing and visualizing trajectories from all kinds of molecular dynamics or Monte Carlo simulations.
MDANSE (Molecular Dynamics Analysis for Neutron Scattering Experiments) is a Python application designed for computing properties that can be directly compared with neutron scattering experiments [GitHub].
MDAnalysis is an object-oriented Python library to analyze trajectories from molecular dynamics simulations.

Visualization and plotting tools

Visualization for Electronic and Structural Analysis (VESTA) program.
OVITO: A scientific visualization and analysis software for atomistic and particle simulation data.
Speck (web-based): High-quality atomistic system rendering program.
CrysX-3D: A crystal visualizer tool available for Windows, MacOS, Linux, and Android devices. The visualizer is built using a gaming engine (Unity3d).
Graphical Display Interface for Structures (GDIS) program.
Avogadro: An advanced molecule editor and visualizer.
Igor Pro offers a complete software solution that's powerful yet agile, giving you the power to visualize, analyze and share your data like never before.
sumo: Python toolkit for plotting and analysis of ab initio solid-state calculation data, built on existing Python packages from the solid-state chemistry/physics community [GitHub].
Matplotlib: A comprehensive library for creating static, animated, and interactive visualizations in Python.

Evolutionary algorithms and machine learning-related programs

MTG in-house code: Nature-inspired Algorithm for Robust Atomic structure search (NARA) is an atomic structure exploration framework based on the Firefly algorithm. The name NARA originates from the Korean expression nal-a (날-아), which represents the flying motion of a firefly. It has been developed as a companion package to LLUMYS, a machine learning interatomic potential that provides the "light" of the firefly.
MTG in-house code: Learning Loss informed Uncertainty Aware MLIP from Yonsei and Sydney (LLUMYS) is an E(3)-equivariant graph neural network interatomic potential that learns per-atom uncertainty based on the learning loss concept. It has been developed as a companion package to NARA (a Firefly algorithm based atomic structure optimization framework). The name LLUMYS originates from lumen, meaning "light", symbolizing the glow of a firefly.
The Gaussian Approximation Potential (GAP) code: To fit interatomic potentials and then use them for molecular simulation (including the use of the Smooth Overlap of Atomic Positions (SOAP) kernel) [GAP tutorial].
The RuNNer Neural Network Energy Representation is a Fortran-based framework for the construction of Behler-Parrinello-type high-dimensional neural network potentials.
SIMPLE-NN (SNU Interatomic Machine-learning PotentiaL packagE – version Neural Network) is an open package that constructs Behler-Parrinello-type neural-network interatomic potentials from ab initio data. The package provides an interfacing module to LAMMPS for MD simulations [GitHub].
SevenNet (Scalable EquiVariance Enabled Neural Network) is a graph neural network interatomic potential package that supports parallel molecular dynamics simulations with LAMMPS.
n2p2: A ready-to-use software for high-dimensional neural network potentials in computational physics and chemistry [GitHub].
PiNN is a Python library built on top of TensorFlow for building atomic neural networks (ANNs) [GitHub].
DeePMD-kit is a package written in Python/C++, designed to minimize the effort required to build deep learning-based models of interatomic potential energy and force field and to perform molecular dynamics (MD) [GitHub].
The Atomic Energy NETwork (ænet) package is a collection of tools for the construction and application of atomic interaction potentials based on artificial neural networks (ANN) [GitHub].
CatLearn: An environment for atomistic machine learning in Python for applications in surface science and catalysis (including a machine learning nudged elastic band (ML-NEB) algorithm) [GitHub].
GOFEE: Efficient global structure optimization with a machine-learned surrogate model -- A Python code for performing global structure search, coupling genetic algorithms with machine learning techniques [GitLab].
CALYPSO (Crystal structure AnaLYsis by Particle Swarm Optimization) is an efficient structure prediction method and it requires only chemical compositions for a given compound to predict stable or metastable structures at given external conditions. It can thus be used to predict/determine the crystal structure and design the multi-functional materials.
XtalOpt: Open-source evolutionary algorithm designed to predict crystal structures.

Other auxiliary programs

MKMCXX: A C++ program for constructing microkinetic models. [wiki]
Zacros is a Kinetic Monte Carlo (KMC) software package written in Fortran, for simulating molecular phenomena on catalytic surfaces.
GIBBS2: Quasi-harmonic Debye model for thermodynamics of solids.
icet: A Pythonic approach to cluster expansions (including code to perform Wang-Landau Monte Carlo runs and generate Special Quasirandom Structures, SQS).
Site Occupancy Disorder (SOD) code: A package of programs and tools for modelling site-disordered solids.
Phonopy and Phono3py: Python-based lattice dynamics and phonon analyzer.
Self Consistent Ab Initio Lattice Dynamical method (SCAILD; implemented in the scph code): A method for calculating temperature-dependent phonon spectra self-consistently from first principles.
ALAMODE is a scientific software designed for analyzing lattice anharmonicity and lattice thermal conductivity of solids. By using an external DFT package such as VASP and Quantum ESPRESSO, you can extract harmonic and anharmonic force constants straightforwardly with ALAMODE. Using the calculated anharmonic force constants, you can also estimate lattice thermal conductivity, phonon linewidth, and other anharmonic phonon properties from first principles.
Sternheimer GW: Many-body perturbation theory without empty states.
multiX: Program to calculate the energy levels of an atom in a crystal field defined by the charges and positions of its neighbours, and to plot the resulting XAS and RIXS spectra.
pyscal: A python module for the calculation of local atomic structural environments (including Steinhardt’s bond orientational order parameters) during post-processing of atomistic simulation data.
Rigorous Investigation of Networks Generated using Simulations (RINGS): A Fortran90/MPI code developed to analyze the results of molecular dynamics simulations, using ring statistics to analyze connectivity.
Local-Orbital Basis Suite Towards Electronic-Structure Reconstruction (LOBSTER): A crystal orbital Hamilton population (COHP) that studies bonding and antibonding contributions to the band-structure energy, and shows the contribution of a specific chemical bond to the band energy. The integrated COHP also hints towards the bond strength of the chemical bond in question.
Z2Pack is a tool for calculating topological invariants. The method is based on tracking the evolution of hybrid Wannier functions, which is equivalent to the computation of the Wilson loop. Originally developed for calculating ℤ2 invariants, it is now also capable of calculating Chern numbers. Moreover, through the use of individual Chern numbers it can be used to identify any kind of topological phase.
WannierTools: Use tight binding model to get the surface states of slab systems or edge states of nanowire systems or just bulk bands. Especially usefull for topological novel systems, including topological insulator, Dirac semimetal, Weyl semimetal, nodal line systems, nodal chain systems, triple point systems and unknown topological systems.
Boltzmann Transport Properties II (BoltzTraP2): A modern implementation of the smoothed Fourier interpolation algorithm for electronic bands that can be used for the calculation of thermoelectric transport coefficients as functions of temperature and chemical potential in the rigid-band picture. [GitLab]
GOLLUM is a program that computes the charge, spin and thermal transport properties of multi-terminal nano-scale junctions. The program can compute transport properties of either user-defined systems described by a tight-binding Hamiltonian, or more material-specific properties of systems composed of real atoms described by DFT Hamiltonians. GOLLUM now interfaces with plane wave codes such as VASP via the Wannier90 code.
NanoTCAD ViDES: A python module, which integrates the C and Fortran subroutines, to simulate nanoscale devices, through the self-consistent solution of the Poisson and the Schrodinger equations, by means of the Non-Equilibrium Green’s Function (NEGF) formalism.
GitHub is an incredibly powerful open source code collaboration/development platform and git repository manager. It makes it easier for developers to be developers: To work together, to solve challenging problems, to create the world’s most important technologies. MTG has a GitHub account too!

Databases (for materials science & engineering)

American Mineralogist Crystal Structure Database
This site is an interface to a crystal structure database that includes every structure published in the American Mineralogist, The Canadian Mineralogist, European Journal of Mineralogy and Physics and Chemistry of Minerals, as well as selected datasests from other journals. The database is maintained under the care of the Mineralogical Society of America and the Mineralogical Association of Canada, and financed by the National Science Foundation.
Open Access Crystallography Database
The goal of this project is to provide 3D visualizations of crystal structures and morphologies in order to help educating future materials scientists and engineers worldwide. This website is also used for research purposes by the Nanocrystallography Group and in class room demonstrations of introductory materials science and engineering courses at Portland State University (PSU). All collected data are freely available over the internet.
Materials Project
By computing properties of all known materials, the Materials Project aims to remove guesswork from materials design in a variety of applications. Experimental research can be targeted to the most promising compounds from computational data sets. Researchers will be able to data-mine scientific trends in materials properties. By providing materials researchers with the information they need to design better, the Materials Project aims to accelerate innovation in materials research.
Open Quantum Materials Database (OQMD)
The OQMD is a database of DFT-calculated thermodynamic and structural properties. This online interface is for convenient, small-scale access; for a more powerful utilization of the data, we recommend downloading the entire database and the API for interfacing with it.
Materials Cloud
Materials Cloud is built to enable the seamless sharing and dissemination of resources in computational materials science, offering educational, research, and archiving tools; simulation software and services; and curated and raw data. These underpin published results and empower data-based discovery, compliant with data management plans and the FAIR principles.
Novel Materials Discovery (NOMAD)
The NOMAD Laboratory maintains the largest Repository, for input and output files of all important computational materials science codes. From its open-access data, it builds several Big-Data Services helping to advance materials science and engineering.
Computational Materials Repository (CMR)
The approach of the Computational Materials Repository (CMR) is to convert data to an internal format that maintains the original variable names without insisting on any semantics. Imported data can be implicitly grouped by user criteria and therefore maintain their natural connection in the database as well. Automatic data analysis is enabled through agents that analyze and group data based on predefined rules. Small projects can be handled without the need of database software while bigger projects one can use to improve performance.
NRELMatDB
NRELMatDB is a computational materials database with the specific focus on materials for renewable energy applications including, but not limited to, photovoltaic materials, materials for photo-electrochemical water splitting, thermoelectrics, etc. The main goal of NRELMatDB is to enable and facilitate the access and exchange of computational data between different research groups following the guidelines outlined in the Presidential Materials Genome Initiative
Bilbao Crystallographic Server
Initiated in 1997, at the Materials Laboratory of the University of the Basque Country, Spain, the Bilbao Crystallographic Server has since been offering its crystallographic and solid state programs and utilities, free of charge.
TOP500 project
The TOP500 project ranks and details the 500 most powerful non-distributed computer systems in the world. The project was started in 1993 and publishes an updated list of the supercomputers twice a year.

Page updated

Google Sites

Report abuse