IoT Data Processing at the Edge with Named Data Networking

Processing high volumes of raw Internet of Things(IoT) data at the network edge is becoming a popular solution toguarantee lower latency interactions compared to the traditionalcomputing in the remote cloud. The synergy between network-ing and computing domains is today enabled by innovativeparadigms such as Named Data Networking (NDN) and its recentextensions, which support the request of computing services “byname” and their distributed execution right inside the networknodes, instead of being deployed in purpose-built servers.In this paper, we extend the NDN architecture to turn thenetwork edge into a dynamic computing environment for runninguser applications relying on IoT data streams processing andanalytics. Novel naming and forwarding mechanisms are definedto properly guide service requests towards edge computing nodes,as close as possible to the IoT data sources, in order to offer lowlatency and avoid that raw IoT data flood the network. Throughsimulations with ndnSIM, we analyse the performance of ourproposal in terms of volume of data traffic and service deliverytime.