Testing Desktop Virtualization
by Angela Miller
Today we kicked off our desktop virtualization test at NCTD. I have a few reasons for wanting to look at virtualization – the first is that the workspaces for some of our users are too crowded and we would like to provide them more space within their constraints without requiring a larger footprint on their desk. This would keep us from needing to expand the offices into additional spaces.
The second compelling reason is to try to extend the lifetime of some of our desktop investment. I previously discussed my choice to upgrade user machines with memory and larger hard disks in an attempt to get at least 2 more years out of the machines. My hope is that desktop virtualization will allow us to contemplate an operating system upgrade without needing to replace every machine at the office.
Our test environment consists of an HPBlade 460, 3 small HP thin clients, 1 HP 6910p laptop, and 2 HP 5100dct desktops. We acquired a VMWare VDI bundle for a relatively small investment and will run our test with these tools. Our plan is to test a variety of elements: stability of the desktop environment, ease of deploying new desktops, the performance of the desktops with a variety of virtual server builds, and performance on the different types of hardware.
We will also be analyzing and researching the power requirements for each of these scenarios. Our hope is that we might see a reduced power requirement even on the older hardware because the server will be doing the ‘heavy lifting’ on the desktop.
After our test, I’ll post the different results we see, including an analysis of whether pursuing a desktop virtualization strategy is ‘greener’ than continuing with our current approach.
Dig deeper on the issues:
I relied on the following sites for analysis in support of this post: