Novel Gossip-Based Multicast Protocol with No Global View Information

Application-level gossip-based broadcast protocols are gaining significant popularity as an interesting alternative to deterministic broadcast algorithms for reliable propagation by providing an attractive tradeoff between reliability and scalability. In addition, many distributed applications require scalable, reliable and causally ordered delivery of broadcast messages to a large number of processes. However, despite the importance of the ordering guarantees, there exist only a few works to consider this requirement. Most early versions of these protocols rely on the assumption that each process knows every other process, or use virtual synchrony probabilistically by composing the gossip-style dissemination based on partially randomized individual views and K-committee protocol with a deterministic reliability. This paper presents an efficient gossip-based multicast protocol to guarantee causally-ordered delivery semantics based on the local view of every individual member consisting of a subset of members continuously changing, whose size never excesses a determined threshold, without virtual synchrony and K-committee.