Multi-run validation of the elevated-ambient operating boundary

Establish through rigorous multi-run experiments whether the exploratory boundary-probe results accurately characterize the ambient-temperature limit at which the empirically calibrated passive DVFS scheduler fails because of nonlinear leakage and thermal throttling.

Background

The reported ambient boundary was inferred from single exploratory probes at 23, 27, and 31 °C. Those measurements suggest that the passive operating envelope closes at approximately 27 °C, but the limited replication prevents reliable characterization of variability, statistical confidence, and the precise failure boundary under elevated ambient conditions.

References

The boundary probes in Table~\ref{tab:boundary} were only conducted once for exploratory purposes and require rigorous multi-run validation in future work.

— Sustainable Edge Vision via Empirically Calibrated DVFS: Eliminating Thermal Throttling on Passively Cooled Hardware  (2609.04705 - Marasini et al., 4 Sep 2026) in Section 7, Limitations; Table 1, “Exploratory boundary-probe runs”