LLMpediaThe first transparent, open encyclopedia generated by LLMs

BigDFT

Note: This article was automatically generated by a large language model (LLM) from purely parametric knowledge (no retrieval). It may contain inaccuracies or hallucinations. This encyclopedia is part of a research project currently under review.
Article Genealogy
Parent: NWChem Hop 5 terminal

This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.

BigDFT
NameBigDFT
DeveloperCEA (French Alternative Energies and Atomic Energy Commission)
Released2008
Programming languageFortran (programming language), C (programming language), Python (programming language)
Operating systemLinux, macOS
GenreComputational chemistry, Density functional theory
LicenseCeCILL

BigDFT BigDFT is a software package for large-scale electronic structure calculations based on density functional theory. It targets simulations of molecules, solids, and materials using a wavelet basis to achieve systematic convergence, combining numerical efficiency with parallel scalability for high-performance computing environments such as TeraGrid and PRACE-class systems.

Overview

BigDFT implements Kohn–Sham Density functional theory methods for ground-state and excited-state properties, emphasizing a multiresolution approach with Daubechies wavelets. The code is designed to run on massively parallel platforms including installations at Barcelona Supercomputing Center, CEA, and other HPC centers, supporting studies related to catalysis, nanotechnology, and condensed matter physics. Development brings together contributions from research groups associated with institutions like Université Grenoble Alpes, CNRS, and industrial partners in computational materials science.

Methodology

BigDFT employs Daubechies wavelets for a systematic, localized basis that offers adaptive resolution in real space, contrasting with plane-wave approaches used in packages tied to VASP, Quantum ESPRESSO, or ABINIT. The implementation incorporates pseudopotentials such as Hartwigsen–Goedecker–Hutter pseudopotentials and techniques for total-energy minimization akin to algorithms found in Car–Parrinello molecular dynamics and iterative eigensolvers like those used in LOBPCG. Exchange–correlation functionals are drawn from the family developed by researchers associated with John Perdew and collaborators, enabling generalized gradient approximations and hybrid functionals comparable to implementations in Gaussian (software) and NWChem.

Implementation and Features

The codebase is written in Fortran (programming language) and C (programming language), with Python bindings for workflow integration and scripting comparable to toolchains used by Atomic Simulation Environment. BigDFT provides features including systematic grid refinement, multigrid Poisson solvers similar to methods in FEniCS Project, and linear-scaling algorithms inspired by work from Stephan Goedecker and Giulia Galli. It supports periodic and non-periodic boundary conditions, implicit solvation models analogous to those employed in VASPsol workflows, and interfaces for molecular dynamics comparable to functionality in LAMMPS.

Performance and Benchmarks

BigDFT demonstrates strong parallel scaling on distributed-memory systems using MPI and hybrid MPI+OpenMP strategies, with benchmark comparisons often presented alongside results from Quantum ESPRESSO, CP2K, and Siesta. Performance assessments highlight efficient memory usage and favorable scaling for thousands of atoms when run on supercomputers such as Oak Ridge Leadership Computing Facility systems and European Tier-0 resources. Optimizations leverage libraries and standards like BLAS, LAPACK, and ScaLAPACK, with profiling and tuning practices comparable to those employed at Argonne National Laboratory.

Applications

BigDFT has been applied to studies in surface chemistry relevant to Haber process catalysis, nanostructure electronic properties connected to Graphene research, defect physics in materials akin to investigations of Silicon (element), and spectroscopy simulations comparable to work on Transition metal complexes. The package supports research workflows in computational materials design pursued by academic groups at institutions such as École Polytechnique, University of Cambridge, and industrial research labs focusing on energy materials and heterogeneous catalysis.

Development History

Initial development traces to research projects at CEA and collaborations with European partners, with early releases appearing around 2008 and subsequent major updates adding hybrid functional support and improved parallel algorithms. The project has interacted with European framework programs and national research initiatives, and development contributors include scientists affiliated with CNRS, Université Grenoble Alpes, and partner universities. Community contributions and shared-development practices mirror collaborative models used by projects like MDAnalysis and CP2K.

Licensing and Availability

BigDFT is distributed under an open-source license compatible with French public-sector software distribution models, and binaries or source packages are available for installation on common HPC platforms. Users can obtain the code through channels maintained by project maintainers and contribute via typical open-source collaboration mechanisms similar to repositories hosted by organizations like GitHub and GitLab.

Category:Computational chemistry software Category:Density functional theory