The advancements in the 5G cellular technology and Industrial IoT enable scenarios with millions of connected data sources. However, these large deployments provide data sources with heterogeneous characteristics, for example, different vendors, versions, and data formats. Therefore, it is arduous for data delivery systems to access and collect data from such scenarios.
The goal of this thesis is to design an adaptive system that can manage the data source interfaces dynamically. This system helps organizations to fetch data from data sources added on-the-fly. A prototype is implemented on top of the Nokia Bell Labs publish-subscribe proof of concept (NBL PoC).
The proposed prototype increases updatability and maintainability of the system, as new requirements from data sources can be applied at runtime on the interfaces. Organizations can quickly deploy and update data source interfaces, without disrupting other active data streams in the NBL PoC.