In recent years, the Internet of Things (IoT) hastransformed objects of everyday life into communicating devices. The number ofconnected devices will be between 10 and 12 billion by 2021 1 ,making itimpossible for current network technologies to support this enormous growth.Future networked systems must adapt existing network architectures to futureneeds and design and develop new management capabilities to help meet thestringent requirements of future use cases. In fact, the upcoming 5G networksaim to tackle these new business opportunities by introducing very high carrierfrequencies, an enormous number of antennas and new functionalities, such asDevice-to-Device communication (D2D) and fog computing 2.Smart cities, counties and urban corridors havestarted to implement the Internet of Things (IoT) platform infrastructures, aswell as the data orchestration and edge computing that are needed to keep upwith the increasing flow of data from different sources. Not only are lightposts a valuable source of real estate for IoT deployments, but they alsorepresent a communications hub for fiber or wireless rollouts. A redundantcommunications infrastructure will be required when the next advent of smartmachines, autonomous vehicles and bots use street fixtures to collect andcalculate, for example, the vehicles’ locations and data sources related todriving, such as traffic congestion and traffic accidents.
In the last few years, 7 IoT applications have beenimplemented as a set of small and independent micro-services. Themicro-services architecture is a relatively new term in software patterns Themicro-service paradigm is an extension of the traditional Service-OrientedArchitecture (SOA) paradigm, where an application is decomposed into a set offine-grained services. Each service communicates through lightweightcommunication protocols. Research studies have been carried out to solve theissues of abstracting end device functionalities, trying to provide a suitablearchitecture with service management and composition capabilities able to linka set of micro-services in a set of IoT applications. Each micro-service can beprovided by a lightweight container, which may be used by multiple tenants.
Ina smart city scenario, 8 resources should be distributed within the networkensuring that the micro-services that make up an application are allocated andinstantiated close to the end device that is requesting the IoT application.Multiple factors should be taken into account to ensure proper resource allocationsuch as latency, bandwidth, energy efficiency and cost.In this article, Internet of Things based managementand orchestration framework is proposed to deal with the application serviceplacement problem in smart cities. Our approach follows the guidelines of theEuropean Telecommunications Standards Institute (ETSI) Network FunctionVirtualization (NFV) Management and Orchestration (MANO) architecture,extending it with additional software components, which will offer not onlyhigh computing performance, but also monitoring and data analysis functionalities.