You could set up monitoring of the VMs & host and determine if you have, or are approaching, a performance bottleneck. Also, if you do want increase capacity, it would be better adding a second node, so you have some redundancy in your lab.