StorPool (unlike RAID), is a self healing system. When a drive or a server fails, StorPool detects the failure and automatically makes new copies of the data, in order to comply with given data protection policies. It will also do automatic load and capacity balancing. If the failed server appears online again, StorPool runs a data recovery process, which updates only the pieces of data which have changed whilst the server was down (and not all of the data). It will also stream the different missing pieces of data from all of the other servers in the cluster, which makes the data recovery process much faster.

StorPool is so flexible that you can even take the drives of a failed server and plug them into a new server, or even place them in empty slots, in servers that are already part of the running cluster and this will also run the data recovery and balancing process, to ensure maximum redundancy and reliability of the system.

StorPool always puts different copies of data in different nodes, in order to have the maximum possible level of redundancy. Indeed if the customer has the scale and should wish to do so, this policy can also be implemented at rack level.