Architecture-Based Mapping Algorithm for Multi-Core Quantum Computers

Abstract

Current quantum processors must be scaled to match the computational power needed for quantum algorithms that present quantum advantage. It is not feasible to keep integrating more qubits on a single core because qubit crosstalk and decoherence make the computations too noisy. Multi-core approaches have been proposed as a feasible, scalable alternative. Optimizing the mapping of quantum algorithms onto multi-core quantum computers is crucial for improving performance and scalability in quantum computing systems. This thesis presents a pipeline for computing the complete mapping of quantum algorithms into multi-core quantum computers, considering both intra-core and inter-core constraints. The existing state-of-the-art inter-core mapping algorithms are evaluated and found to be suboptimal for achieving near-optimal inter-core communications. An alternative inter-core mapping algorithm is proposed to address this limitation, demonstrating improved effectiveness in allocating quantum operations across multiple cores. A thorough comparison of the proposed algorithm and existing approaches is conducted using the complete mapping pipeline. The findings emphasize the importance of developing efficient inter-core mapping algorithms. The research outcomes contribute to advancing the field of quantum computing and provide valuable insights for optimizing performance and scalability in multi-core quantum computers.