Tensor-structured algorithm for reduced-order scaling large-scale Kohn–Sham density functional theory calculations

We present a tensor-structured algorithm for efficient large-scale density functional theory (DFT) calculations by constructing a Tucker tensor basis that is adapted to the Kohn–Sham Hamiltonian and localized in real-space. The proposed approach uses an additive separable approximation to the Kohn–S...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:npj computational materials Jg. 7; H. 1; S. 1 - 9
Hauptverfasser: Lin, Chih-Chuen, Motamarri, Phani, Gavini, Vikram
Format: Journal Article
Sprache:Englisch
Veröffentlicht: London Nature Publishing Group UK 12.04.2021
Nature Publishing Group
Nature Portfolio
Schlagworte:
639/301/1034/1037
639/301/1034/1038
Algorithms
Characterization and Evaluation of Materials
Chemistry and Materials Science
Computational Intelligence
Density functional theory
Localization
Materials Science
Mathematical analysis
Mathematical and Computational Engineering
Mathematical and Computational Physics
Mathematical Modeling and Industrial Mathematics
Scaling
Tensors
Theoretical
ISSN:2057-3960, 2057-3960
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We present a tensor-structured algorithm for efficient large-scale density functional theory (DFT) calculations by constructing a Tucker tensor basis that is adapted to the Kohn–Sham Hamiltonian and localized in real-space. The proposed approach uses an additive separable approximation to the Kohn–Sham Hamiltonian and an L 1 localization technique to generate the 1-D localized functions that constitute the Tucker tensor basis. Numerical results show that the resulting Tucker tensor basis exhibits exponential convergence in the ground-state energy with increasing Tucker rank. Further, the proposed tensor-structured algorithm demonstrated sub-quadratic scaling with system-size for both systems with and without a gap, and involving many thousands of atoms. This reduced-order scaling has also resulted in the proposed approach outperforming plane-wave DFT implementation for systems beyond 2000 electrons.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2057-3960
2057-3960
DOI:10.1038/s41524-021-00517-5