Agreed with the general lack of algorithm experience. However, this knowledge can still be obtained professionally. I think, if you can't say "look at my degree", you should be able to say "look at all this software I've built over the past four years, it proves I know a thing or two".
I haven't actually dumped an open source application in favour of a proprietary alternative because it was difficult to install. I could, however, completely understand why some users after attempting to install and configure open source applications would make the move. The installation and initial configuration experience is the first impression and is important to get right. It is the first impression the software gives as to how good/usable/stable it really is. How can you get feedback regarding other aspects of the software if they can't get past step one?
The virt-top command line utility proves quite useful for monitoring virtual machine load. For any kind of automated task that should take place as a result of this load, the same data can be retrieved via libvirt as that is what virt-top uses.