Conclusion

Dispatching is often essential to control and guarantee time constraints in complex real-time systems. But such dispatching leads to new problems that need to be solved. One such problem is real-time scheduling. Indeed, respecting the global temporal constraints of a distributed real-time application implies respecting both the temporal constraints of the tasks and those of the messages exchanged between these tasks. We have summarized the various techniques that enable the placement and migration of real-time tasks and the communication of messages exchanged between these tasks in distributed architectures.

We have emphasized that placement can be static or dynamic. Static placement is easy to implement and facilitates scheduling tests. Unfortunately, it does not offer the flexibility to react to site overloads, or to take certain dynamic aspects into account....