Performance evaluation of interoperable micro-clouds

The Internet of Things (IoT) is defined as a paradigm transforming physical objects to smart objects that are inter-connected via Internet. Today, IoT objects offer embedded intelligence that can be powerful in case of fully integration of a collective manner towards the satisfaction of user needs. This work is based on the micro-clouds that are a new proposing paradigm to highlight the collective intelligence of IoT objects. Specifically, a micro-cloud could be seen as a pool of cooperated devices and their resources that form transient smart environments. Further to this, we anticipate that the inter-cloud model can expand the micro-cloud capabilities by allowing multiple micro-clouds to communicate in order to achieve a common aim. This will further push the boundaries for studying the interaction and synergetic collaborative nature between micro-cloud systems in terms of their interoperability and performance. However as the size of the system is increased the complexity of performance is additionally increased. This emphasizes the need for decentralization where resources are changing over time without any notice. The vision of this work is that micro-clouds shall be linked together to enable a full network of usable IoT objects and at the same time maintain the required quality of service from an end-user's perspective. Specifically, the aim is to identify the specific criteria which are the most relevant to optimize performance when several micro-clouds collaborate (e.g. load-balancing, throughput, turn-around times, utilization level, etc.) as well as classify their functional requirements. So the focus is on the performance analysis and evaluation of results based on a simulated specific use case scenario.