The Consensus Problem with Fallible Communication in Cloud Computing Environment

Distributed systems are growing rapidly in nowadays. Under many circumstances, reaching a common agreement in the presence of faulty components is the central issue of fault tolerant distributed computing. The cloud computing that can make more application for user from the Internet. The new concept, cloud computing, it is using the low-power hosts to achieve high reliability that will ensured the ability to be better. Consensus protocols, which allow the healthy nodes to agree on a common value, have been brought up to aid the reliable execution of tasks. The cloud computing that can make more application for user from the Internet, so that, the high capability of fault-tolerant in cloud computing environment needs to be considered. Traditionally, the consensus problem was visited in a fully connected network, broadcasting network and generalize connected network. However, the topology of cloud computing is not a certainty structure. Therefore, in this study, the consensus problem is re-examined in a topology of cloud computing. The proposed protocol, Consensus Protocol for Fallible Link (CPFL), can make all healthy nodes reaching consensus with minimal number of message exchanges and tolerant the maximal number of allowable faulty components. The protocol that we proposed attempts to solve the consensus problem and makes all healthy nodes in the distributed system can get stable results without any influence from faulty components.