We present a complete linear scaling method for hybrid Kohn-Sham density functional theory electronic structure calculations and demonstrate its performance. Particular attention is given to the linear scaling computation of the Kohn-Sham exchange-correlation matrix directly in sparse form within the generalized gradient approximation. The described method makes efficient use of sparse data structures at all times and scales linearly with respect to both computational time and memory usage. Benchmark calculations at the BHandHLYP/3-21G level of theory are presented for polypeptide helix molecules with up to 53 250 atoms. Threshold values for computational approximations were chosen on the basis of their impact on the occupied subspace so that the different parts of the calculations were carried out at balanced levels of accuracy. The largest calculation used 307 204 Gaussian basis functions on a single computer with 72 GB of memory. Benchmarks for three-dimensional water clusters are also included, as well as results using the 6-31G** basis set.

Density matrix purification, although being a powerful tool for linear scaling construction of the density matrix in electronic structure calculations, has been limited by uncontrolled error accumulation. In this article, a strategy for the removal of small matrix elements in density matrix purification is proposed with which the forward error can be rigorously controlled. The total forward error is separated into two parts, the error in eigenvalues and the error in the occupied invariant subspace. We use the concept of canonical angles to measure and control differences between exact and approximate occupied subspaces. We also analyze the conditioning of the density matrix construction problem and propose a method for calculation of interior eigenvalues to be used together with density matrix purification.

We present a second-order recursive Fermi-operator expansion scheme using mixed precision floating point operations to perform electronic structure calculations using tensor core units. A performance of over 100 teraFLOPs is achieved for half-precision floating point operations on Nvidia’s A100 tensor core units. The second-order recursive Fermi-operator scheme is formulated in terms of a generalized, differentiable deep neural network structure, which solves the quantum mechanical electronic structure problem. We demonstrate how this network can be accelerated by optimizing the weight and bias values to substantially reduce the number of layers required for convergence. We also show how this machine learning approach can be used to optimize the coefficients of the recursive Fermi-operator expansion to accurately represent the fractional occupation numbers of the electronic states at finite temperatures.

An accelerated polynomial expansion scheme to construct the density matrix in quantum mechanical molecular dynamics simulations is proposed. The scheme is based on recursive density matrix expansions, e.g., [A. M. N. Niklasson, Phys. Rev. B, 66 (2002), 155115], which are accelerated by a scale-and-fold technique [E. H. Rubensson, J. Chem. Theory Comput., 7 (2011), pp. 1233-1236. The acceleration scheme requires interior eigenvalue estimates, which may be expensive and cumbersome to come by. Here we show how such eigenvalue estimates can be extracted from the recursive expansion by a simple and robust procedure at a negligible computational cost. Our method is illustrated with density functional tight-binding Born-Oppenheimer molecular dynamics simulations, where the computational effort is dominated by the density matrix construction. In our analysis we identify two different phases of the recursive polynomial expansion, the conditioning and purification phases, and we show that the acceleration represents an improvement of the conditioning phase, which typically gives a significant reduction of the computational cost. Introduction.With the fast growth of computational processing power, atomistic simulations based on calculations of the electronic structure have become a powerful approach to the study of a broad range of problems in materials science, chemistry, and biology [26,33,64]. Nevertheless, the computational cost associated with electronic structure calculations normally limits applications to fairly small systems. In particular, the cubic, O(N 3 ), scaling of the computational cost as a function of the number of atoms, N , for the regular solution of the quantum mechanical eigenvalue problem is considered to be a most limiting factor. A number of different electronic structure technologies have therefore been developed that circumvent this bottleneck with a computational effort that scales only linearly with the system size [6,18]. The reduction in the cost is typically achieved by utilizing sparse matrix algebra in an iterative construction of the density matrix, which avoids the full regular solution of the quantum mechanical eigenvalue problem. The matrix sparsity arises in localized atomic basis set representations due to the short-range character of the electronic wavefunctions for nonmetallic materials [3,27,28]. With linear scaling