Validation of an Analysability Model in Hybrid Quantum Software

TL;DR

This work tackles the challenge of analysability in hybrid quantum software, where classical and quantum components create unique maintainability hurdles. It proposes a hardware-agnostic analysability model that blends classical code metrics with quantum circuit metrics, implemented through a SonarQube plug-in and a Power BI visualization platform, and validated via an international controlled experiment. Key contributions include the specification of properties and metrics (e.g., $NC\_CW1$, $DC\_CW1$), the integrated tooling environment, and empirical validation across institutions. The study aims to advance Quantum Software Engineering by providing a practical framework to assess and improve the understandability and modifyability of hybrid software systems.

Abstract

In the context of quantum-classical hybrid computing, evaluating analysability, which is the ease of understanding and modifying software, presents significant challenges due to the complexity and novelty of quantum algorithms. Although advances have been made in quantum software development, standard software quality evaluation methods do not fully address the specifics of quantum components, resulting in a gap in the ability to ensure and maintain the quality of hybrid software products. In this registered report proposal, we intend to validate a quality model focused on the analysability of hybrid software through an international collab orative approach involving academic institutions from Italy and Spain through a controlled experiment. This approach allows for a more detailed analysis and validation methodology and establishes a framework for future research and developments in software quality assessment in quantum computing.

Figures (3)

• Figure 1: Methodology overview.
• Figure 2: Technological environment.
• Figure 3: Results visualization tool.