The BigDFT-suite Package

BigDFT is a suite containing many different programs, all of which are maintained on our gitlab page Instead of building the code with one single configure line, the code is built as a suite of different packages.

buildprocedure.png

In this scheme we might see how the BigDFT code itself is separated. The figure describes the interdependencies among these packages. Let us describe each of the packages which are depicted here. we have here separated the packages into upstream contributions and native contributions.

We start by describing the packages that are part of the internal development efforts.

Native packages

  • futile: a library handling most common FORTRAN low-level operations like memory management, profiling routines, I/O operations. It also supports yaml output and parsing for fortran programs. It further provides wrappers routines to MPI and linear algebra operations. This library is extensively used in BigDFT packages;

  • atlab: A library which is associated to the description of the operation on atomic structures. It contains the handling of the structural information and manipulations of it. It should be agnostic of the DFT formalism of the code employing it.

  • CheSS: A module for performing Fermi Operator Expansions via Chebyshev Polynomials, released as a separate project on Launchpad

  • psolver: a flexible real-space Poisson Solver based on Interpolating Scaling Functions. It constitutes a fundamental building block of BigDFT code, and it can also be used separately and linked to other codes. It also internally uses the futile library for the I/O.

  • libABINIT: this is a subsection of files coming from ABINIT software package, to which BigDFT has been coupled since the early days. It handles different parts like symmetries, ewald corrections, PAW routines, density and potential mixing routines and some MD minimizers. Also some XC functionals, initially natively implemented in the ABINIT code, are also coded in this library. This library uses the futile code through the (experimental) PAW section.

  • BigDFT: the core routines for computing electronic structure;

  • spred: a library for structure prediction tools that is compiled on top of BigDFT routines.

Upstream packages

  • libyaml: this library is used to parse the yaml markup language that is used in BigDFT input files;

  • PyYaml: this is a Python module which makes it possible to convert yaml into python objects. This part is mainly used for postprocessing purposes as BigDFT logfile also comes in yaml format;

  • libXC: this library handles most of the XC functionals that can be invoked from BigDFT runs;

  • GaIn: this library handles analytic integrals of common operators between Gaussian Functions. It does not perform low-level operations and can be linked separately;

Optional packages

Sometimes on large supercomputers it can be difficult to manage and install supplementary packages. For this reason, we offer a number of optional packages that can be activated by the installer. The full set of options can be found in the modulesets directory. This directory regroups the information about the packages which can (or have to) be installed before the native bigdft suite packages. Here a non exhaustive list:

  • ase: a python package for running DFT calculations with different codes.

  • biopython: a python package for bioinformatics.

  • dill: a python package for serializing complex objects.

  • fftw: fast fourier transform.

  • netlib: linearl algebra operations (better to use system’s libraries)

  • openmm: for classical molecular dynamics simulations.

  • openbabel: for chemical informatics.

  • rdkit: another python code for chemical informatics.

  • sirius: a DSL for electronic structure calculations.

  • spglib: a library for handling crystal symmetries.

  • wannier90: for computing maximally localized Wannier functions.

Build System

In the older versions of the code, all these different packages were compiled with the same configuration instructions. With the present version, each of the code sections described above can be considered as a separate package, which improves modularity between code sections and reduces side-effects. In addition, each package can now be compiled with different installation instructions.

We are using a build suite tool based on the Jhbuild project which is regularly employed by developers of gnome project. We have re-adapted/added some of the functionality of the jhbuild package to meet the needs of our package.