Efficient Circuit Cutting and Scheduling in a Multi-Node Quantum System with Dynamic EPR Pairs

2024-12-24

AI Summary

To address fidelity degradation and high classical reconstruction overhead in large-scale quantum circuit execution on multi-node quantum systems—caused by qubit limitations, hardware noise, and circuit cutting—the paper proposes a synergistic execution framework integrating dynamic EPR-pair generation with adaptive circuit cutting. It introduces the first dynamic mapping mechanism from logical to physical EPR pairs, minimizing the number of subcircuits and associated classical communication costs. Furthermore, it formulates a unified noise-aware and hardware-resource-constrained optimization model for joint circuit cutting and scheduling. The framework is implemented in Qiskit, supporting graph-based partitioning, dynamic EPR allocation, parallel subcircuit execution across nodes, and tensor-network-based state reconstruction. Experimental evaluation on real quantum hardware and multiple simulators demonstrates up to 16.7% fidelity improvement and a 99.5% reduction in total system overhead compared to Qiskit-Addon-Cut.

Technology Category

• Machine Learning: Quantum Machine Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Distributed Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Despite advancements, current quantum hardware faces significant challenges, including limited qubit counts and high susceptibility to noise, which hinder the execution of large, complex algorithms. To address these limitations, multi-node quantum systems and quantum circuit cutting techniques partition large circuits into smaller subcircuits that can be executed on individual quantum machines and then reconstructed using classical resources. However, these methods introduce new challenges, such as the large overhead from subcircuit reconstruction and additional noise from entangled EPR pairs, especially in multi-node quantum systems. In this paper, we propose the Efficient Circuit Cutting and Scheduling (EC2S) system, which integrates EPR pairs with circuit cutting to address these issues. EC2S improves system performance by transitioning from logical to physical EPR pairs and further reduces computational overhead by minimizing the number of subcircuits during the reconstruction phase. sol~ is implemented using Qiskit and evaluated on both real quantum hardware and various emulators. Compared to the state-of-the-art Qiskit-Addon-Cut, EC2S achieves significant improvements in fidelity, up to 16.7%, and reduces system-wide expenditure by up to 99.5%.

Problem

• Quantum Computing
• Algorithm Execution
• Quantum Bits Limitations

Innovation

• EC2S
• Multi-Quantum Computer Coordination
• Circuit Partitioning Optimization