CONQUEST is a large scale DFT electronic structure code, capable of both diagonalisation and linear scaling, or O(N), calculations. It is developed jointly by NIMS (National Institute for Materials Science, Japan), ISM (Institut des Science Moleculaires, University of Bordeaux) and UCL. The code is designed to perform DFT calculations on very large systems (containing tens of thousands, hundreds of thousands or even millions of atoms) though is also very fast for smaller systems, and scales exceptionally well with numbers of processes. It can be run at different levels of precision, ranging from ab initio tight binding up to full DFT with plane wave accuracy. It is capable of operation on a range of platforms from workstations up to high performance computing centres. These web pages contain information on the code, and its applications, as well as separate areas for developers.
CONQUEST is now available as an open source project under an MIT licence. A recent comprehensive overview has been published: J. Chem. Phys. 152, 164112 (2020); a freely available copy is on arXiv.
The code is presently on v1.1, with v1.2 to be released in July 2023. It is a complete, robust code which is compatible with a very accurate library of pseudopotentials, PseudoDojo. We welcome bug reports and suggestions for improvement through the GitHub issues page. We are very happy to welcome new users and developers.
If you are interested in more details about O(N) methods, you can find a comprehensive review which we wrote here (external link), also available on arXiv.
You can find out more about Conquest in these pages:
A roadmap for electronic structure codes on exascale computers
CONQUEST is used to provide ab initio MD data for improved fitting of machine learning forces
A review of multi-site support functions (MSSFs) in CONQUEST
CONQUEST was used to study anionic gold nanoparticle catalysts
CONQUEST is used to provide ab initio MD data for fitting machine learning forces