Practical Continuous Aggregation in Wireless Edge Environments

Abstract

The edge computing paradigm brings the promise of overcoming the practical scalability limitations of cloud computing, that are a result of the high volume of data produced by Internet of Things (IoT) and other large-scale applications. The principle of edge computing is to move computations beyond the data center, closer to end-user devices where data is generated and consumed. This new paradigm creates the opportunity for edge-enabled systems and applications, that have components executing directly and cooperatively on edge devices. Having systems' components, actively and directly, collaborating in the edge, requires some form of distributed monitoring as to adapt to variable operational conditions. Monitoring requires efficient ways to aggregate information collected from multiple devices. In particular, and considering some IoT applications, monitoring will happen among devices that communicate primarily via wireless channels. In this paper we study the practical performance of several distributed continuous aggregation protocols in the wireless ad hoc setting, and propose a novel protocol that is more precise and robust than competing alternative.