Scalable Simulation of Hierarchical Quantum Systems on Quantum and Classical Processors
Wednesday, June 24, 2026 3:45 PM to 5:15 PM · 1 hr. 30 min. (Europe/Berlin)
Foyer D-G - 2nd Floor
Research Poster
Emerging Computing TechnologiesParallel Numerical AlgorithmsQuantum Program Development and OptimizationQuantum Computing - Use CasesSimulating Quantum Systems
Information
Poster is on display and will be presented at the poster pitch session.
Hamiltonian simulation - the computation of quantum time evolution - is a central challenge in scientific computing and a key domain where quantum processors are expected to deliver transformative impact. Yet, efficiently simulable Hamiltonians belong to restricted structural classes on both quantum and classical hardware. Exploiting such structure is therefore essential for quantum algorithm development, particularly given the current limitations of quantum devices in coherence time and qubit count, which make large-scale classical simulation indispensable for validation and benchmarking.
We introduce a unified framework for simulating networks of interacting quantum subsystems, termed Hierarchical Quantum Systems (HQS). Such structures arise in molecular aggregates, clustered network models, and constrained combinatorial optimisation. Although HQS Hamiltonians may appear complex, we show that when local subsystem dynamics and global inter-subsystem coupling admit efficient representations, the full system admits an exact decomposition into independently evolvable subspaces. This yields substantial reductions in computational complexity on both quantum and classical platforms.
For quantum processors, we construct circuits that exactly simulate HQS dynamics with only constant-factor overhead relative to the corresponding reduced global dynamics. For classical platforms, the same structural insight leads to a scalable distributed algorithm in which global communication scales with the number of subsystems rather than the full Hilbert space dimension, while local subspace evolution is naturally parallel.
Building on these results, we present QSpace, an open-source software framework that unifies quantum circuit generation and MPI-distributed classical simulation under a common interface. QSpace enables compositional modelling of hierarchical quantum walks and supports execution across gate-based quantum hardware and large-scale HPC systems. Benchmarks on the Setonix HPE Cray EX supercomputer demonstrate near-linear strong scaling to thousands of MPI ranks with minimal communication overhead. We further illustrate applications in modelling light-harvesting dynamics and combinatorial optimisation in logistics.
Together, these results establish hierarchical structure as a unifying principle for scalable quantum simulation and provide a practical bridge between quantum algorithm design and HPC-scale verification.
Hamiltonian simulation - the computation of quantum time evolution - is a central challenge in scientific computing and a key domain where quantum processors are expected to deliver transformative impact. Yet, efficiently simulable Hamiltonians belong to restricted structural classes on both quantum and classical hardware. Exploiting such structure is therefore essential for quantum algorithm development, particularly given the current limitations of quantum devices in coherence time and qubit count, which make large-scale classical simulation indispensable for validation and benchmarking.
We introduce a unified framework for simulating networks of interacting quantum subsystems, termed Hierarchical Quantum Systems (HQS). Such structures arise in molecular aggregates, clustered network models, and constrained combinatorial optimisation. Although HQS Hamiltonians may appear complex, we show that when local subsystem dynamics and global inter-subsystem coupling admit efficient representations, the full system admits an exact decomposition into independently evolvable subspaces. This yields substantial reductions in computational complexity on both quantum and classical platforms.
For quantum processors, we construct circuits that exactly simulate HQS dynamics with only constant-factor overhead relative to the corresponding reduced global dynamics. For classical platforms, the same structural insight leads to a scalable distributed algorithm in which global communication scales with the number of subsystems rather than the full Hilbert space dimension, while local subspace evolution is naturally parallel.
Building on these results, we present QSpace, an open-source software framework that unifies quantum circuit generation and MPI-distributed classical simulation under a common interface. QSpace enables compositional modelling of hierarchical quantum walks and supports execution across gate-based quantum hardware and large-scale HPC systems. Benchmarks on the Setonix HPE Cray EX supercomputer demonstrate near-linear strong scaling to thousands of MPI ranks with minimal communication overhead. We further illustrate applications in modelling light-harvesting dynamics and combinatorial optimisation in logistics.
Together, these results establish hierarchical structure as a unifying principle for scalable quantum simulation and provide a practical bridge between quantum algorithm design and HPC-scale verification.
Format
on-demandon-site