Runtime Load-Shifting of Distributed Controllers Across Networked Devices

The ability to monitor and steer the behaviour of complex distributed systems is an increasingly hot research topic, fostered by the growing adoption of hybrid cloud-edge technologies that constitute a computational continuum. One key feature of these systems is the ability to scale in size, embracing a wide number of heterogeneous devices and applications. This complexity, in turn, impacts the monitoring and control systems that need, at the same time, to be able to deal with high complexity and computational load and be available on all kinds of devices. In this paper, we introduce an architecture that allows for shifting the computational load of monitor systems at runtime across different devices in the cloud-edge continuum. We show the feasibility of the proposed approach by providing a reference implementation integrated with an existing simulation platform, leveraging Kotlin multiplatform to allow interoperability among different runtimes.