Low communication high performance ab initio density matrix renormalization group algorithms

Abstract

There has been recent interest in the deployment of ab initio density matrix renormalization group computations on high performance computing platforms. Here, we introduce a reformulation of the conventional distributed memory ab initio DMRG algorithm that connects it to the conceptually simpler and advantageous sum of sub-Hamiltonians approach. Starting from this framework, we further explore a hierarchy of parallelism strategies, that includes (i) parallelism over the sum of sub-Hamiltonians, (ii) parallelism over sites, (iii) parallelism over normal and complementary operators, (iv) parallelism over symmetry sectors, and (v) parallelism over dense matrix multiplications. We describe how to reduce processor load imbalance and the communication cost of the algorithm to achieve higher efficiencies. We illustrate the performance of our new open-source implementation on a recent benchmark ground-state calculation of benzene in an orbital space of 108 orbitals and 30 electrons, with a bond dimension of up to 6000, and a model of the FeMo cofactor with 76 orbitals and 113 electrons. The observed parallel scaling from 448 to 2800 CPU cores is nearly ideal.

14 pages, 6 figures; typos corrected

References in corpus (10)

• Heat-bath Configuration Interaction: An efficient selected CI algorithm inspired by heat-bath sampling
• Semistochastic Heat-bath Configuration Interaction method: selected configuration interaction with semistochastic perturbation theory
• Low-energy spectrum of iron-sulfur clusters directly from many-particle quantum mechanics
• The density matrix renormalization group for ab initio quantum chemistry
• A spin-adapted Density Matrix Renormalization Group algorithm for quantum chemistry
• Multireference Correlation in Long Molecules with the Quadratic Scaling Density Matrix Renormalization Group
• Generic Construction of Efficient Matrix Product Operators
• Real-Space Parallel Density Matrix Renormalization Group
• A General Automatic Method for Optimal Construction of Matrix Product Operators Using Bipartite Graph Theory
• Investigation of metal-insulator like transition through the ab initio density matrix renormalization group approach

Cited by in corpus (31)

• Reliably assessing the electronic structure of cytochrome P450 on today's classical computers and tomorrow's quantum computers
• Block2: a comprehensive open source framework to develop and apply state-of-the-art DMRG algorithms in electronic structure and beyond
• The chromium dimer: closing a chapter of quantum chemistry
• Systematic electronic structure in the cuprate parent state from quantum many-body simulations
• Initial state preparation for quantum chemistry on quantum computers
• Matrix product states with large sites
• Density Matrix Renormalization Group with Tensor Processing Units
• Variational Lang-Firsov approach plus Møller-Plesset perturbation theory with applications to ab initio polariton chemistry
• Rapid initial state preparation for the quantum simulation of strongly correlated molecules
• Quantum Information-Assisted Complete Active Space Optimization (QICAS)
• Quasi-Lindblad pseudomode theory for open quantum systems
• Interacting models for twisted bilayer graphene: a quantum chemistry approach
• Hierarchical Clifford transformations to reduce entanglement in quantum chemistry wavefunctions
• Accurate Simulation of the Hubbard Model with Finite Fermionic Projected Entangled Pair States
• A comparison between the one- and two-step spin-orbit coupling approaches based on the ab initio Density Matrix Renormalization Group
• Fault-tolerant quantum computation of molecular observables
• Tensor network influence functionals in the continuous-time limit: connections to quantum embedding, bath discretization, and higher-order time propagation
• Conservation laws in coupled cluster dynamics at finite-temperature
• Accuracy of ghost-rotationally-invariant slave-boson theory for multiorbital Hubbard models and realistic materials
• Multireference protonation energetics of a dimeric model of nitrogenase iron-sulfur clusters
• Impact of conditional modelling for a universal autoregressive quantum state
• Interpolating many-body wave functions for accelerated molecular dynamics on the near-exact electronic surface
• Fermionic tensor network contraction for arbitrary geometries
• Piecewise Interaction Picture Density Matrix Quantum Monte Carlo
• Electronic specific heat capacities and entropies from density matrix quantum Monte Carlo using Gaussian process regression to find gradients of noisy data
• In-Depth Investigation of Phase Transition Phenomena in Network Models Derived from Lattice Models
• The Software Landscape for the Density Matrix Renormalization Group
• Quantum max-flow in the bridge graph
• Quantum Many-Body Linear Algebra, Hamiltonian Moments, and a Coupled Cluster Inspired Framework
• Chemical bonding concepts emerge naturally from maximally entangled atomic orbitals
• Enhanced Krylov Methods for Molecular Hamiltonians: Reduced Memory Cost and Complexity Scaling via Tensor Hypercontraction