Enhancing Fault Tolerance of Cloud Nodes using Replication Techniques

Abstract

The cloud technologies are gaining boom in the field of information technology. But on the same side cloud computing sometimes results in failures. These failures demand more reliable frameworks with high availability of computers acting as nodes. The request made by the user is replicated and sent to various VMs. If one of the VMs fail, the other can respond to increase the reliability. A lot of research has been done and being carried out to suggest various schemes for fault tolerance thus increasing the reliability. Earlier schemes focus on only one way of dealing with faults but the scheme proposed by the the author in this paper presents an adaptive scheme that deals with the issues related to fault tolerance in various cloud infrastructure. The projected scheme uses adaptive behavior during the selection of replication and fine-grained checkpointing methods for attaining a reliable cloud infrastructure that can handle different client requirements. In addition to it the algorithm also determines the best suited fault tolerance method for every designated virtual node. Zheng, Zhou,. Lyu and I. King (2012).