Companies now see massive operating cost savings in their Data Centres with virtualization. They have less physical hardware, use less power, and can provision new servers and services within minutes. This is all good and well, but how do we ensure that what we deploy in our virtualized infrastructures is replicated in real time and then in the event of a failure is failed over without incurring a second of down time? VMware’s answer to that question is Fault Tolerance. VMware Fault Tolerance, is the only feature available today that can guarantee the uptime of virtual machines in your environment. It provides zero downtime, zero data loss and continues availability for your applications even in the event of an ESX host hardware failure, and that without the cost and complexity of traditional hardware and software clustering solutions. VMware is the only vendor that provides capabilities and features such as Fault Tolerance, and is therefore the only virtualization vendor that can actually guarantee the ongoing availability of your critical applications.
Unlike VMware HA (High Availability) which restarts virtual machines on other ESX hosts with available capacity, Fault Tolerance eliminates downtime altogether. Fault Tolerance is based on leading edge technology called vLockstep and eliminates even the smallest of data loss, corruption or disruption to service. Fault tolerance can protect any virtual machine, enabling a single and simple technology to protect all applications. This enables applications that cannot be protected by traditional clustering methods to be protected. As Fault Tolerance can easily be “turned on” or “turned off” for individual virtual machines in your existing HA clusters, Fault Tolerance provides a simple solution to the complex problems that surrounds redundant systems and clustering.
How it works:
Fault Tolerance uses the well known encapsulation properties of virtualization by building high availability directly into the x86 hypervisor in order to deliver hardware style fault tolerance to virtual machines. It requires no custom hardware or software (no need to purchase expensive software editions such Enterprise Edition for applications such as SQL, Exchange and even operating systems such as Windows Server Enterprise Edition). VMware Fault Tolerance relies on VMware vLockstep technology to establish and maintain an active secondary virtual machine that runs in virtual lockstep with the primary virtual machine. The secondary virtual machine resides on a different host and executes exactly the same sequence of guest instructions as the primary virtual machine. The secondary VM observes the same input as the primary VM and is ready to take over at any time without any data loss or interruption of service should the primary VM fail. Both VMs are managed as a single unit, but run on different physical ESX hosts.
Enabling Fault Tolerance does not require any additional configuration changes to the virtual machine’s guest OS or its applications. It's enabled on an existing virtual machine with a few clicks. This eliminates the need to purchase enterprise edition licenses for your corporate operating systems and applications. As it’s enabled on individual Virtual Machines, you can choose to protect only the applications that are mission critical. You can even run the primary and secondary virtual machines in different buildings, providing an instant DR solution.
In the event that the primary virtual machine becomes unavailable, the secondary virtual machine takes over with no interruption to service. After the initial failover, VMware HA automatically selects a new host that is operating properly and has available resource and starts a new secondary virtual machine instance on that host. The hypervisor where new primary virtual machine is running establishes vLockstep with the new secondary hypervisor, thus re-enabling redundancy. From this point onward, the virtual machine is protected one more against future failures.