Are you asking about high availability clustering (fail-over clusters) of individual instances running in a cloudstack managed environment?
If so, there are numerous products that provide that capability for most Linux distros,
RHEL having the most well documented one I'm familiar with.