Epidemic-style Causal Order Broadcasting Only Using Partial View

Because network-level reliable group communication protocols rely on IP multicast and have lack of reliability, this motivates the demand on application-level group communication. Epidemic-style protocols among these application-level approaches guarantee reasonably high reliability, provide good scalability and are easy to deploy. But, earlier versions of these protocols often rely on the assumption that every process knows every other process. There exist epidemic-style broadcast protocols based on partially randomized individual views providing totally ordered delivery of broadcast messages to members of large groups. However, there is no such epidemic-style broadcast protocol providing causally ordered delivery property, which is very useful for many distributed applications such as video-conferencing and multi-party games. This paper proposes an efficient epidemic-style broadcast protocol to guarantee causally ordered delivery semantics based on the local view of every individual member consisting of a subset of members, which continuously evolves, but never exceeds a fixed size.