Energy-Efficient QoS-Aware Application and Network Configuration for Next-Gen IoT.

The growth of the Internet of Things (IoT) paradigm has brought the attention from intensive domains, with the objective of automating their critical processes in what is known as the next-generation IoT. Nonetheless, nextgeneration IoT applications will be critical, and thus, will have very high Quality of Service (QoS) requirements, like very short response times. To achieve these short response times, the research community proposes using the Cloud Continuum, as well as Software-Defined Networking (SDN). Moreover, IoT applications are often based on microservices architectures due to their distributed nature, enabling them to fully obtain the benefits of the Cloud Continuum. Therefore, placing the next-generation IoT application’s microservices and the SDN controllers in a manner that minimizes the response times is key. However, using the Cloud Continuum and SDN with the sole objective of minimizing response times will incur in a higher energy consumption and, by proxy, an even larger carbon footprint due to the implementation of next-generation IoT. In this paper, we present Q-NGINE, a framework that enables an energy-efficient microservice and SDN controller placement through the Cloud Continuum, trading off the energy consumption of the next-generation IoT application and its response time. An evaluation over an Internet of Medical Things use case shows that Q-NGINE can reduce the energy consumption up to 24.19% with an increase of only 1.5% in response times.