Massively Parallel Tensor Network State Algorithms on Hybrid CPU-GPU Based Architectures

TL;DR

This paper introduces advanced parallel tensor network algorithms optimized for hybrid CPU-GPU architectures, enabling large-scale quantum simulations with unprecedented Hilbert space dimensions.

Contribution

It presents novel algorithmic solutions and implementation strategies that significantly extend the capabilities of tensor network state algorithms on high-performance computing systems.

Findings

Achieved large-scale DMRG simulations on Hilbert spaces up to 2.88×10^36 dimensions.

Demonstrated the effectiveness of hybrid CPU-GPU architectures for tensor network computations.

Extended the limits of quantum many-body simulations using novel parallel algorithms.

Abstract

The interplay of quantum and classical simulation and the delicate divide between them is in the focus of massively parallelized tensor network state (TNS) algorithms designed for high performance computing (HPC). In this contribution, we present novel algorithmic solutions together with implementation details to extend current limits of TNS algorithms on HPC infrastructure building on state-of-the-art hardware and software technologies. Benchmark results obtained via large-scale density matrix renormalization group (DMRG) simulations are presented for selected strongly correlated molecular systems addressing problems on Hilbert space dimensions up to $2.88\times10^{36}$.