E.L. Dominguez, J. Estudillo Ramirez, J. Fanchon, and S.E. Pomares Hernandez (Mexico)
Distributed systems, causal order, causal distance, immediate dependency relation, forward error recovery, unreliable networks.
In this work we propose an efficient real-time causal broadcast algorithm with fault tolerance to unreliable networks. The algorithm allows for the delivery of causal messages with recovering capabilities on real time systems by using the concept of redundancy. Redundancy, in our work, is calculated based on the causal distance. The concept of causal distance was first introduced to detect the immediate dependency relation (IDR,) which presents a distance equal to one. We extend the causal distance idea in order to add minimal redundancy information (d>1). With information redundancy we ensure the causal delivery property even in the presence of lost messages. Our algorithm is suitable to be used in real-time systems because it has the characteristic of forward error recovery.
Important Links:
Go Back
