Smart Resource Planning for Live Migration in Edge Computing for Industrial Scenario

Future factory automation systems are expected to process vast amounts of data and orchestrate complex cyber-physical components. This involves devices such as autonomous guided vehicles and augmented reality glasses that are inherently mobile. Edge Computing (EC) is a promising approach to address the requirements set by upcoming industrial systems. However, the combination of distributed edge network and multiple mobile devices gives rise to the need for live migration. Migration of multiple services in parallel demands for an intelligent allocation of network resources. This paper describes a dynamic algorithm to perform smart allocation of resources for live migration on demand. In addition to maintaining the Quality of Service requirements of the end devices, it also maintains the network overhead due to live migration low. We evaluate the performance of our algorithm using a discrete-event network simulator.