GPU-Accelerated Quantum Circuit Simulation in the Pauli Propagation Framework via cuQuantum SDK

The Pauli path propagation method is a recent and promising addition to the toolbox of efficient classical algorithms for simulating large quantum circuits. By describing a quantum system as an approximate evolution of quantum operators in the Heisenberg picture by a truncated Pauli path integral, expectation values of much larger quantum systems can be efficiently computed. Unfortunately, CPU-oriented algorithms based on the tree search are unable to fully leverage the power of modern massively parallel GPU architectures. To alleviate this problem, cuQuantum SDK now offers GPU-accelerated building blocks for efficient implementation of the Pauli path propagation methods on NVIDIA GPUs. The provided low-level APIs give simulator developers a fine-grain control over the specific Pauli path propagation protocol they are interested in implementing, while ensuring its efficient execution on NVIDIA GPUs. Furthermore, with the power of multiple GPUs, users can now scale up their simulations to even larger quantum systems than previously possible.