The DynaRoute use case depicts a realistic situation of a person following a predefined "itinerary" (i.e. a scheduled sequence of activities) on her way from her hotel to the airport. All intermediate activities involve the dynamic synthesis and adaptation of choreographies which are controlled and modified on-the-fly, based on external triggers and condition changes. DynaRoute involves People, Things and Services of the Future Internet, harmoniously coordinated together in a distributed fashion in order to achieve multiple goals. Choreographies are adapting within a very dynamic environment in a scalable and efficient manner.
The scenario, which is considered in the DynaRoute use-case, is depicted in the following figure:
Figure 5: The DynaRoute use case scenario (where MICE stands for Mobile DevICE)
The actors in the above scenario (and corresponding choreographies) are:
What distinguishes this use case is the fact that it utilizes mostly local, bidirectional communications between various actors, as well as (localized) location-based services (LBS). In this way, thousands of actors (people, things or services) can interact with each other in a variety of ways, maintaining low service complexity, faster response times and truly feasible scalability. Alternatively, mobile users would have to spend valuable, costly and limited global bandwidth in order to access centralized Internet servers. Moreover with our distributed approach, local information, such as population density, can trigger local reaction in a more direct and responsive way, while alike global changes most often need time to be registered, validated and propagated through.
The scalability of our proposed method comes from the fact that these are localized services, delivered in a distributed manner, in "cells" of 300m-500m radius (assumed range of the mobile device), without centralized bottlenecks, which scale poorly to hundreds of thousands of users over large geographical areas. CHOReOS will generalize and prove by simulation and analytical methods the scalability of this distributed choreography approach to thousands of users per cell, given the actual performance figures of intelligent mobile devices and of the entire choreography framework, which will be recorded during the demonstration.