We focused on functional testing since performance testing between such different systems wouldn’t have been useful. We set up volumes with thin provisioning, with initial capacities of 20GB and total capacities of 100GB, then added data to ensure that the volumes could automatically expand when they reached the capacity threshold for the original volume, generally around 80% full. This worked properly for all systems tested.
We focused on functional testing since performance testing between such different systems wouldn’t have been useful. We set up volumes with thin provisioning, with initial capacities of 20GB and total capacities of 100GB, then added data to ensure that the volumes could automatically expand when they reached the capacity threshold for the original volume, generally around 80% full. This worked properly for all systems tested.
We then set up snapshots of the original volume, made changes, created a second snapshot, then mounted the original snapshot as a separate volume and used it to restore the changed files to their original state. We also tested the ability to take a snapshot of a virtual hard disk and mount the snapshot as a new VM using both VMware and hyper-V.
This worked well with both platforms. We also tested the functionality of replication. We had access to a remote Compellent system to test remote replication, and Dell EqualLogic sent us two systems so we could test local replication. In the other cases, we went through the process of creating local and remote replicas, looking at the difficulty of configuring replication.
