Scalable electron correlation methods I.: PNO-LMP2 with linear scaling in the molecular size and near-inverse-linear scaling in the number of processors.

摘要: We propose to construct electron correlation methods that are scalable in both molecule size and aggregated parallel computational power, the sense total elapsed time of a calculation becomes nearly independent molecular when number processors grows linearly with size. This is shown be possible by exploiting combination local approximations algorithms. The concept demonstrated linear scaling pair natural orbital second-order Moller–Plesset perturbation theory (PNO-LMP2) method. In this method, wave function manifold integrals transformed incrementally from projected atomic orbitals (PAOs) first orbital-specific virtuals (OSVs) finally (PNOs), which allow for minimum domain sizes fine-grained accuracy control using very few parameters. A algorithm design discussed, efficient small large molecules, numbers processors, although tru...