Smart SDN Management of Fog Services

Abstract

<p>We present a smart Service Manager whose role is</p> <p>to direct user requests (such as those coming from IoT devices)</p> <p>at the edge towards appropriate servers where the services they</p> <p>request can be satisfied, when services can be housed at different</p> <p>Fog locations, and the system is subject to variations in workload.</p> <p>The approach we propose is based on using an SDN controller as</p> <p>a decision element, and to incorporate measurement data based</p> <p>machine learning that uses Reinforcement Learning to make the</p> <p>best choices. The system we have developed is illustrated with</p> <p>experimental results on a test-bed in the presence of time-varying</p> <p>loads at the servers. The experiments confirm the ability of the</p> <p>system to adapt to significant changes in system load so as to</p> <p>preserve the QoS perceived by end users.</p>