Efficient multicasting agreement protocol

Abstract

The fault-tolerance is an important research topic in the study of distributed systems. To cope with the influence from faulty components, reaching a common agreement at the presence of faults before performing some special tasks is essential. Therefore, the Byzantine agreement (BA) problem has drawn more and more of the researchers' attention as they explore deeper and deeper into the world of distributed systems. Traditionally, the BA problem has been visited in a Fully Connected Network (FCN), BroadCasting Network (BCN) or Generalize Connected Network (GCN), making the malicious fault assumption with processors or communication media grows into the dual failure mode on both processors and communication media. However, the network structures (topologies) of FCN, BCN and GCN are not practical nowadays. In this study, we shall take a close look at the limits of the network structure to revisit the BA problem with a Multi-Casting Network (MCN). FCN, BCN and GCN are all special cases of MCN, so the new protocol we shall propose here will also be able to solve the BA problem with FCN, BCN and GCN structures. Our new protocol uses a minimum number of rounds of message exchange and can tolerate a maximum number of allowable faulty components and make each fault-free processor reach a common agreement in an MCN.