The service could be improved by faster servers, more widely available VMs, and more storage in one place. Then, in the event of a blade failure, we could start our VMs on another blade in a couple of minutes. This is why we wanted to buy storage. But now, we want to buy a twin server with 24 VMware discs to create test storage. The next product release should allow more servers to be controlled simultaneously.