Mind the Shift: Secure Migration of Containerized Processes in Edge Computing

Edge Computing is a new technology with increasing relevance for various areas of application. In particular, the industrial IoT domain is driving demands for efficient data analytics, fast reaction to results of data evaluation, and the requirement to adapt to dynamic changes in and of the environment. Computing tasks and operations need to be efficient, their management automated as much as possible. Virtualization technologies allow for effective and scalable utilization of such computing resources. This paper addresses the sub-domain of dynamic distribution of containerized processes at runtime over virtualization environments. This process migration is enabled in a way that running processing tasks are stopped, the state gets externalized, the data is transferred to another location and resumed there eventually. Our contribution focus on security. It proposes to protect such live process migration to avoid misuse and risks to the integrity of the processing tasks despite migration.We present the attacker model based threats, motivate the proposed security model, discuss our implementation and present its validation result.