Linear-Scaling ab-initio Calculations for Large and Complex Systems

Linear-scaling ab initio calculations: applications and future directions

Ab initio calculations for large dielectric matrices of confined systems.

Calculations for optical excitations in confined systems require knowledge of the inverse screening dielectric function epsilon(-1)(r,r(')), which plays a crucial role in determining exciton binding energies. We present a new efficient real-space method of inverting and storing large ab initio dielectric matrices of confined systems, which relies on the separability of epsilon matrix in r and r...

Large-scale ab initio quantum chemical calculations on biological systems.

In this Account we describe recent advances in two ab initio electronic structure methods, localized perturbation approaches and density functional theory, that allow accurate calculations including electron correlation to be carried out for systems with hundreds of atoms. Application of these methods to large-scale modeling of biological systems is discussed. Localized perturbation methods are...

Parallel scaling of Teter’s minimization for Ab Initio calculations

We propose a parallelization scheme for the conjugate gradient method by Teter et. al. and report a detailed analysis of its scalability. We use MPI collective operations exclusively to take advantage of optimized collective implementations with possible hardware support. Our parallel conjugate gradient calculation can be applied in addition to the already implemented parallelism in the applica...

Enabling ab initio Hessian and frequency calculations of large molecules.

A linear scaling method, termed as cardinality guided molecular tailoring approach, is applied for the estimation of the Hessian matrix and frequency calculations of spatially extended molecules. The method is put to test on a number of molecular systems largely employing the Hartree-Fock and density functional theory for a variety of basis sets. To demonstrate its ability for correlated method...