VFT: A virtualization and fault tolerance approach for cloud computing

Abstract

Fault tolerance in cloud computing is a grand challenge problem now a days. The main fault tolerance issues in cloud computing are detection and recovery. To combat with these problems, many fault tolerance techniques have been designed to reduce the faults. In this paper, we have proposed a Virtualization and Fault Tolerance (VFT) technique to reduce the service time and to increase the system availability. A Cloud Manager (CM) module and a Decision Maker (DM) are used in our scheme to manage the virtualization, load balancing and to handle the faults. In the first step virtualization and load balancing are done and in the second step fault tolerance is achieved by redundancy, checkpointing and fault handler. VFT is mainly designed to provide reactive fault tolerance. In our proposed approach a fault handler is included in the virtualization part. Fault handler blocks the unrecoverable faulty nodes along with their virtual nodes for future requests and removes the temporary software faults from the recoverable faulty nodes and makes them available for the future requests.