Establish real-time feasibility on accelerated edge hardware for faster control cycles

Determine whether an accelerated edge node can satisfy the complete BALLAST control-plane decision-to-enforcement latency budget when the industrial control cycle is 100 milliseconds rather than 500 milliseconds.

Background

The paper measures the complete decision-to-enforcement path only on the workstation. It measures Digital Twin inference separately on a Jetson GPU, where inference fits the 50-millisecond budget derived from the SWaT scan cycle, but does not measure the complete control-plane path on that platform.

For faster industrial systems with a 100-millisecond scan cycle, the allowable 10 percent budget is only 10 milliseconds. The paper explicitly states that the workstation does not meet this budget and that available measurements are insufficient to determine whether accelerated edge hardware would meet it.

References

At the fast end of the SCADA range, a 100\,ms cycle, the 10 percent budget falls to 10\,ms. The complete path was measured only on the workstation, where it costs 21.9\,ms in the mean, so that budget is not met there, and we do not have the control-plane measurements to say whether an accelerated edge node would meet it.

— BALLAST: Physics-Aware Moving Target Defense for Industrial Control Systems Using a Digital Twin Safety Oracle  (2610.18311 - Mogollon-Gutierrez et al., 13 Oct 2026) in Section 5.2, subsection “Latency and Real-Time Feasibility”