Availability Modeling and Analysis on Virtualized Clustering with Rejuvenation

Summary Nowadays, more stringent compliance regulations, businesses of all sizes are required to implement security measures to ensure their systems and data are readily available and quickly recoverable. Virtualized clustering is inherently designed for maximum data integrity and minimal recovery time in the case of site failure. Virtualization allows multiple operating system instances to run concurrently on a single physical machine. In this paper, we present a technique that can increase availability of application servers through the use of virtualization, clustering and software rejuvenation. Using analytical modeling, we analyze multiple design choices when a single physical server and dual physical servers are used to host multiple virtual machines. Analysis results are included to show the performance of the proposed method. By integrating virtualization, clustering and software rejuvenation, it is possible to benefit from increased availability, manageability and savings from server consolidation through virtualization without decreasing uptime of critical services.