Acceleration of Atomistic NEGF: Algorithms, Parallelization, and Machine Learning
This work enables more efficient quantum transport simulations for nanoscale device design, though it appears to be an incremental compilation of existing algorithmic improvements.
This paper addresses the computational challenges of simulating quantum transport in nanoscale devices using DFT+NEGF methods, summarizing algorithmic advances that enable simulations of realistic system sizes and discussing how machine learning could further accelerate these simulations.
The Non-equilibrium Green's function (NEGF) formalism is a particularly powerful method to simulate the quantum transport properties of nanoscale devices such as transistors, photo-diodes, or memory cells, in the ballistic limit of transport or in the presence of various scattering sources such as electronphonon, electron-photon, or even electron-electron interactions. The inclusion of all these mechanisms has been first demonstrated in small systems, composed of a few atoms, before being scaled up to larger structures made of thousands of atoms. Also, the accuracy of the models has kept improving, from empirical to fully ab-initio ones, e.g., density functional theory (DFT). This paper summarizes key (algorithmic) achievements that have allowed us to bring DFT+NEGF simulations closer to the dimensions and functionality of realistic systems. The possibility of leveraging graph neural networks and machine learning to speed up ab-initio device simulations is discussed as well.