2026-10-11T22:08:20Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33715780 2026-09-13T17:48:52Z portal_1197 item_type_3 month_year_09_2026

Data for A universal preconditioner for simulating condensed phase materials David Packwood (7600676) James R. Kermode (14201739) Letif Mones (1900060) Noam Bernstein (1595503) John Woolley (18791658) Nicholas Gould (3318177) Christoph Ortner (11648294) Gabor Csanyi (9242925) Condensed matter -- Computer simulation Mathematical optimization Migrated from ePrints We introduce a universal sparse preconditioner that accelerates geometry optimisation and saddle point search tasks that are common in the atomic scale simulation of materials. Our preconditioner is based on the neighbourhood structure and we demonstrate the gain in computational efficiency in a wide range of materials that include metals, insulators and molecular solids. The simple structure of the preconditioner means that the gains can be realised in practice not only when using expensive electronic structure models but also for fast empirical potentials. Even for relatively small systems of a few hundred atoms, we observe speedups of a factor of two or more, and the gain grows with system size. An open source Python implementation within the Atomic Simulation Environment is available, offering interfaces to a wide range of atomistic codes.<br><br>## Required software - ASE (currently `jameskermode` fork needed, will shortly be merged into ASE trunk): git clone https://gitlab.com/jameskermode/ase cd ase git checkout precon python setup.py install - Jupyter notebook: pip install jupyter - QUIP/quippy (optional, for some of the interatomic potentials): git clone https://github.com/libAtoms/QUIP cd QUIP make config make make quippy make install-quippy - MatSciPy (optional, for fast neighbour lists): git clone https://github.com/libAtoms/matscipy cd matscipy python setup.py install - PyAMG (optional, for fast preconditioner inversion/application): pip install pyamg - Some of the tests also require VASP, CASTEP or CP2K DFT codes - Some of the notebooks require Julia (v0.4.x) ## General files README-dataset.txt - this file notebooks/plots.ipynb - Jupyter notebook (Python) ## Data files for Fig. 1 notebooks/spectra3.ipynb - Jupyter notebook (Julia) ## Data files for Fig. 2 Si_slab/quip_params.xml - SW force field parameters Si_slab/test_precon.xyz - input structure Si_slab/results.new.C1_Exp_3.0_Pfrommer_ID.json - results (all cases) Si_slab/run.py - Python script ## Data files for Fig. 3 Si_crack/params.xml - SW force field parameters Si_crack/crack.xyz - input structure Si_crack/crack.json - results (ID, A=0, A=3 preconditioners) Si_crack/crack_Pfrommer.json - results (Pfrommer preconditioner) Si_crack/test-crack.ipynb - Jupyter notebook ## Data files for Fig. 4 notebooks/precon-scaling.ipynb - Jupyter notebook (Python) ## Data files for Fig. 5 PAW setups used are from the vasp 4.6 distribution: regular La and Al, and O_s (soft). LaAlO3_crack/LaAlO3_crack.c0.xyz - input structure LaAlO3_crack/KPOINTS - VASP kpoint mesh LaAlO3_crack/run.py - Python script LaAlO3_crack/results.ID.json - results LaAlO3_crack/results.C1.json - results LaAlO3_crack/results.Exp_3.0.json - results LaAlO3_crack/results.Pfrommer.json - results ## Data files for Fig. 6 PAW setups used are from the vasp 4.6 distribution: regular Al, but the conventional O. gamma_Al2O3/gamma_Al2O3_Johannes_1.01.c0.xyz 0 input structure gamma_Al2O3/INCAR.template - VASP input file gamma_Al2O3/run.py - Python script gamma_Al2O3/results.C1.json - results gamma_Al2O3/results.Exp_3.0.json - results gamma_Al2O3/results.ID.json - results gamma_Al2O3/results.Pfrommer.json - results ## Data files for Fig. 7 ice/run/run.py - Python script ice/run/common/BASIS_SET - CP2K input ice/run/common/cp2k_input_force.template - CP2K input ice/run/common/iceVIII.xyz - CP2K input ice/run/common/POTENTIAL - CP2K input ice/results/dump.json.VIIIbig.geom.new_preconpy_cp2k_force_preconLBFGS_ID_fmax_0.001_armijo - results ice/results/dump.json.VIIIbig.geom.new_preconpy_cp2k_force_preconLBFGS_Exp_A_0.0_rcut_2.25_fmax_0.001_armijo - results ice/results/dump.json.VIIIbig.geom.new_preconpy_cp2k_force_preconLBFGS_Exp_A_3.0_fmax_0.001_armijo - results ice/results/dump.json.VIIIbig.geom.cell.new_preconpy_cp2k_force_stress_preconLBFGS_ID_fmax_0.001_armijo - results ice/results/dump.json.VIIIbig.geom.cell.new_preconpy_cp2k_force_stress_preconLBFGS_Exp_A_0.0_rcut_2.25_fmax_0.001_armijo - results ice/results/dump.json.VIIIbig.geom.cell.new_preconpy_cp2k_force_stress_preconLBFGS_Exp_A_3.0_fmax_0.001_armijo - results ## Data files for Fig. 8 notebooks/dimer-fig.ipynb - Jupyter notebook (Julia) 2016-04-20T00:00:00Z Dataset Dataset 10.82444/warw.33715780.v1 https://figshare.com/articles/dataset/Data_for_A_universal_preconditioner_for_simulating_condensed_phase_materials/33715780 CC BY 4.0