With StorPool we were able to build live failover on top of our LXC infrastructure. This allows us both to live-migrate containers between compute nodes without any downtime and, in case of an entire node suffering any type of failure, we can bring all containers back online within a minute on a spare compute node.
Creating snapshots within seconds for big disks has helped our different migration projects since it allows us to perform them in a short period of time.
The two 10GE networks provide redundancy and increased performance as they serve as two separate networks doubling the throughput and doing multipathing and load balancing. We now have a high performance shared storage system which enables us to run on private cloud. Our previous system used bare-metal hardware, which provided high performance but inflexible management. Now we have best of both worlds, SSD-class performance with flexibility of a private cloud system.
I have personally met with multiple Storpool engineers and spoke about different options and features. There are too many features that we don't know or use yet. My recommendation would be to promote the new features and give users different examples of how they can be used and how we can benefit from them.
It would be good if, with next releases, StorPool provide a better GUI for monitoring and statistics. This would make our experience even better and complete.
