Could temporal burstiness be mitigated by spreading SWAPs across multiple clock cycles?

# Spreading SWAPs Across Clock Cycles and Temporal Burstiness

Yes, spreading SWAP operations across multiple clock cycles is directly relevant to mitigating temporal burstiness in quantum processors.

"Temporal burstiness" describes a situation where many SWAP gates are concentrated into a narrow time window, creating a spike in resource demand [AG-2024.02-1271]. Think of it like rush-hour traffic: instead of cars spreading evenly across the day, they bunch up, overwhelming the infrastructure.

The key insight is that temporal burstiness arises from the qubit routing problem—when qubits need to be brought together for two-qubit gates on connectivity-constrained hardware, the necessary SWAP sequence can create this bursty execution pattern [AG-2024.02-1271]. By spreading those SWAPs across more clock cycles, you flatten the temporal profile.

This directly maps to circuit compilation strategies discussed in the archive. Several papers show that routing algorithms can be optimized to reduce not just the *total number* of SWAPs but also their temporal concentration [AG-2024.05-2540]. The weighted token swapping algorithm minimizes "SWAP-gate depth"—which is exactly the sequential clustering you're asking about—and simultaneously reduces accumulated error [AG-2024.05-2540].

Why this matters: temporal burstiness can exacerbate error rates because multiple qubits experience high-fidelity operations simultaneously, increasing the risk of correlated errors and reducing the effectiveness of error correction [AG-2024.04-2400]. Spreading SWAPs temporally gives your error-correction codes breathing room.

However, the archive doesn't contain a paper specifically comparing burst-mitigation *via temporal spreading* against other mitigation strategies (like re-mapping or detector-based cutoffs); that focused analysis isn't covered here.