Failover and clustering of SQL Server instances
It allowed us to have redundant servers for our SQL instances so in the event of system failure, we would only be down for a few minutes while things failed over
Failover times are slow. When failover errors occur, can cause problems on entire cluster. Poor logging of issues. Slow support.
When one system failed over, it would sometimes cause IP conflicts resulting in communication between certain servers being interrupted and sometimes causing our SAN to stop communicating with the SQL blades.
No, scalability worked great.
Low. I have found HP support to be a bit lacking whenever I needed it.
Unsure of initial cost. Day to day cost is very low. The system, when functioning properly, requires practically no day to day cost.
PolyServe was good, but hit EOL. I found a suitable alternative. The failover worked, but took too long. And for some odd reason, while failing over, it would take down all of our SQL instances until they were all available. DxConsole was a suitable replacement made by the same guys who developed PolyServe.