Hardware-Aware Performance Characterization of Small Dense Lindblad Propagation for Near-Term Quantum Control

Dense Lindblad propagation at the Hilbert-space sizes of near-term transmon control ($d = 3$, $9$, $27$) sits in a regime where cache boundaries, launch overhead, and fixed latency decide which hardware is useful, not peak arithmetic rate. This paper characterizes that regime on two x86 CPUs, two NV