Event Processing across Edge and the Cloud for Internet of Things Applications

The rapid growth of sensing devices has opened up complex event processing (CEP) for real-time analytics in Internet of Things (IoT) Applications. While CEP has traditionally been centralized, the increasing capabilities of edge devices like smart phones, and the operational needs of low latency and privacy makes it desirable to use both edge and the Cloud for distributed CEP, the former often serving as event sources. This paper motivates the need for real-time analytics across edge and the Cloud, formalizes an optimization problem for bi-partitioning a CEP query pipeline based on IoT application needs, and proposes an initial solution.