As dependencies on computing increase, ensuring that applications are highly available becomes more critical. Many enterprise hardware components have built-in redundancies, such as RAID storage, ECC memory, and multiple power supplies in servers. However, customers are increasingly interested in protecting the entire operating environment from complete hardware failures, and not just the failure of individual components.
Some existing fault-tolerant systems address this requirement, but said systems typically require either proprietary hardware or modified software and usually support only a handful of operating systems. More often, administrators use software-based solutions such as clustering to provide high availability. However, such clustering solutions are difficult to set up and are often tied to specific applications and operating systems.
Read this paper and get VMware fault tolerance recommendations and considerations on VMware vSphere 4.
