A Data Protection Focused Adaptation Engine for Distributed Video Analytics Pipelines

Abstract

The design, development, deployment, and operation of a distributed Video Analytics Pipeline (VAP) at the edge of the network is highly complex. In the domain of adaptive systems, several solutions are proposed in literature to optimize either one particular performance aspect of a VAP, e.g., execution time or latency, or focus on minimal energy consumption, or calculate a trade-off including some of those aspects. However, nowadays, most systems utilizing a VAP that records personally identifiable data have to adhere to some form of data protection regulation, such as the GDPR. Still, adaptations to increase data protection requirements are often second to previously mentioned performance or energy consumption characteristics of a VAP. While there is state of the art literature dealing with data protection related adaptations, most of them solely focus on increasing certain security or privacy aspects of a system, leaving previously mentioned performance or energy consumption characteristics out of scope. To the best of our knowledge, there is no solution that covers all of these aspects. In this paper, we present a data protection focused adaptation engine that leverages the application- and infrastructure based adaptation space of a distributed VAP. The engine employs an extended system model and adaptation rules that are based on previous research. It features an optimization algorithm to improve data protection, performance and energy consumption characteristics of a distributed VAP.

Keywords

• Adaptive systems
• optimization
• video analytics pipelines
• data protection
• edge computing