An Integrated Virtualized Strategy for Fault Tolerance in Cloud Computing Environment
Bashir Mohammed, Mariam Kiran, Irfan‐Ullah Awan, Kabiru M. Maiyama · 2016
Cloud fault tolerance allows the cloud an ability to keep performing its functions correctly even if faults are occurring in the system. This becomes an important property that enables a complete system to continue functioning in the event of one or more faults for high availability of virtual machines or in life critical systems. A fault-tolerant design may allow the system to possibly function at a reduced level, rather than failing completely. As a major concern in guaranteeing availability, reliability of critical services or application execution in cloud environment, cloud fault tolerance research focuses on detection, recovery strategies. However, in order to minimize impacts, anticipate failures to proactively handle them, a model called an Integrated Virtualized Failover strategy (IVFS) was introduced where fault tolerance was realized using redundancy, checkpoint/replay, fault manager. In this paper, we critically analyze this model, proposed a model that tolerate faults based on the reliability of each computing node or virtual machine, removing these from the availability list if the performance is not optimal. The results of our algorithm presents an increase in pass rates, considers forward/backward recovery using diverse software tools. Our simulation results suggest a good performance compared to current existing models. The results are demonstrated through experimental validation with a critical analysis, laying the foundation for a fully fault tolerant IaaS Cloud environment.