I thought it was pretty amazing and failed over very quickly.
I did encounter one big issues with its virtualized disk subsystem and the locking it does on the physical disk.
I had an instance basically get the power plug pulled out by the windows team.
When PolyServe attempted to fail over, polyserve's virtual disk subsystem had a lock still held by the powered down server because that server never released the lock on a clean shutdown.
I ended up having to take down the whole cluster of 6 servers and 12 instances to straiten it out.
That one example withstanding, it was solid and failed over wicked fast.
Patching is a bit of a process, but I've seen worse.
Be cautious about the Achilles heel I mentioned. It was about 2-3 years ago so maybe they've addressed the issue.
Good luck.