State of practice: evaluating GPU performance of state vector and tensor network methods

2024-01-11

arXiv.org

AI Summary

Classical simulation of quantum circuits on NISQ-era hardware faces severe efficiency bottlenecks, particularly when scaling to larger qubit counts and deeper circuits. Method: We systematically benchmark GPU-accelerated state-vector and tensor-network simulation methods across 31 configurations and eight representative NISQ subroutines, quantifying the impact of circuit structure—including entanglement depth and gate density—and qubit count on simulation performance. We propose the first circuit-feature-driven automatic strategy selection criterion, integrating NP-hard contraction-path optimization (formulated as a max-cut problem), tensor-network contraction scheduling, and regression-based performance prediction. Contribution/Results: Our approach achieves up to 10× speedup over baseline methods across diverse NISQ subroutines, precisely delineates the applicability boundaries of state-vector versus tensor-network simulation, and breaks key memory and computational bottlenecks. We publicly release a reusable benchmarking framework and an accurate, generalizable performance prediction model.

Technology Category

• Machine Learning: Matrix & Tensor Methods
• Search and Optimization: Sampling/Simulation-based Search
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

The frontier of quantum computing (QC) simulation on classical hardware is quickly reaching the hard scalability limits for computational feasibility. Nonetheless, there is still a need to simulate large quantum systems classically, as the Noisy Intermediate Scale Quantum (NISQ) devices are yet to be considered fault tolerant and performant enough in terms of operations per second. Each of the two main exact simulation techniques, state vector and tensor network simulators, boasts specific limitations. The exponential memory requirement of state vector simulation, when compared to the qubit register sizes of currently available quantum computers, quickly saturates the capacity of the top HPC machines currently available. Tensor network contraction approaches, which encode quantum circuits into tensor networks and then contract them over an output bit string to obtain its probability amplitude, still fall short of the inherent complexity of finding an optimal contraction path, which maps to a max-cut problem on a dense mesh, a notably NP-hard problem. This article aims at investigating the limits of current state-of-the-art simulation techniques on a test bench made of eight widely used quantum subroutines, each in 31 different configurations, with special emphasis on performance. We then correlate the performance measures of the simulators with the metrics that characterise the benchmark circuits, identifying the main reasons behind the observed performance trend. From our observations, given the structure of a quantum circuit and the number of qubits, we highlight how to select the best simulation strategy, obtaining a speedup of up to an order of magnitude.

Problem

• Quantum Computing Simulation
• Tensor Network Optimization
• Memory Limitations

Innovation

• Quantum Circuit Simulation
• Performance Optimization
• Simulation Speed Enhancement