The Internet of Things (IoT) is expected to be the next paradigm to explode in IT with 50 billion connected devices by 2020.

The real-time requirement and largely distributed nature of “things” may not fit well with the traditional Cloud paradigm. They may favor distributed or hybrid paradigms, where the compute and storage resources span into the edge such as fog or mobile edge computing.

These systems are built on top of devices that, on the one hand, are heterogeneous, constrained, and often exacerbate the challenges in terms of dependability and reliability. On the other hand, the devices are also owned by multiple entities and increasingly pervasive, which heightens the challenges with regard to security and privacy.

We can understand these systems by using complex data analytics techniques and intelligent modelling when choosing the inputs/-outputs and the causality effects. Moreover, to lower the management complexity, these systems need to be self-controlled by autonomic loops and tend to have a high level of local autonomicity.