VMWare Fault Tolerance is a pretty sweet technology…if only you could run it on more than your image server.
VMWare limits Fault Tolerance to a single core
In a recent post VMWare vs RedHat Virtualization I was pretty excited about using VMWare’s Fault Tolerance feature. This allows a VM to run on 2 separate servers with each command being executed in perfect sync across them. Should one fail, the other will pick up without losing a single instruction. It is important to note that non-deterministic events aren’t replicated. After looking very closely into the documentation, they sneak inside the requirements that the VM can only run on a SINGLE CORE. The screen shot listed is under “vSphere Availability Guide” What is even more interesting is the fact that the 8 page document “Protecting Mission-Critical Workloads with VMware Fault Tolerance” from VMWare fails to make a single reference to this limitation.
Hey you know what..many VM’s run just dandy with a single core. Let’s party like its 1999!
You heard that right.. If you run a critical application on 4 cores, 3 cores or even 2 cores, you are out of luck. Apparently they are working on multiple core support in the 12 to 18 month time frame.
