QABE: a Framework for Quantum Annealer Programming and Benchmarking

In recent years, research has given great focus to quantum annealers, a novel architecture belonging to the quantum computing realm, which stands out for its potential to boost computation while being able to tackle more complex problems than gate-based quantum computers currently can. Yet, their programming model requires deep expertise, and estimating their performance is a non-trivial task, due to the limits of current device topologies. Therefore, this work introduces QABE, a framework for automatic programming and benchmarking of quantum annealers through a high-level interface. We describe the software architecture of our solution, that relies on the Quadratic Unconstrained Binary Optimization (QUBO) representation. Furthermore, we select four optimization problems and provide an experimental benchmarking of the D-Wave Advantage architecture, highlighting performance and resource utilization.