My post discussing a VDI pilot project in conjunction with netbooks produced a variety of responses. These readers chose not to post their comments to the public site, so I thought it would be a good idea to highlight two of the most interesting ones.

Licensing, Server Costs, Price Difference of Netbooks & Laptops

One reader asked me, "has the project considered the cost of Microsoft licensing, server costs and the relatively small price difference between laptops and netbooks?" The study that I have in hand from Buch's consultant does factor in the client access licenses required.

Mobility is important to this project, so I thought I'd go out and test this reader's assertion about pricing between laptops and netbooks. To do a quick and dirty sampling, I went to a name-brand PC maker's site last night and got pricing for enterprise laptops versus their netbooks. Enterprise-ready fat laptops started at $1100 and went to $2800. Their netbook offering started at $279 and went up to $499. So, the difference in cost would be at least $821, and up to $2521. Multiply that by a thousand clients, and you can start to see that this is going to pay for servers relatively quickly.

Client management is another story, of course, but you could argue that it is easier to deploy management tools to a netbook that doesn't have much of anything on it -- and only needs things like the operating system & browser patched -- than it would be to deploy management tools to a fat client.

Another Deployment Story

Jim Bates, a technical director at ITAGroup, Inc, an Iowa-based management consulting group wrote to us, saying:

I appreciate your article. In fact, I love sitting in my soft, comfortable leather chair at home with my netbook on my knee accessing my corporate VDI desktop. At work my "netbook" is a 6 year old repurposed desktop PC that just serves as a thin client. It's less expensive to use this dinosaur desktop PC than to purchase a netbook or thin device as a replacement. I would like to have the power savings from such a switch but those savings don't tip the scale in favor of a replacement; at least not yet. ... In my case we were motivated to deploy desktop virtualization by disaster recovery. With 95% of my employees working under a single roof the thought of having to reimage a few hundred traditional PCs after a disaster felt daunting and a little ridiculous. Each day we replicate our virtual desktop personalization settings between data centers. Two weeks ago during our annual disaster test we proved this mechanism worked just fine as employees showed up at our alternate work site and logged into the "same" virtual desktop they left the day before. I've been using VDI for about 6 months and can't even tell the difference between it and the thick desktop/laptop I once used.

That six-year-old end user device should make you sit up and take notice. Extending the lifecycle of a typical corporate desktop to six years is an increase of anywhere from doubling the lifecycle to 50% more of a lifecycle. That can translate into big bucks.

And, maybe the power savings aren't worth it now, but it really depends upon what you are going from and what you are going to. The Niagara VDI project that I discussed in my last blog compared their old way of doing things with desktops to netbooks. comparing a typical desktop at 175kWh with a nebook at 81Kwh. The report that we saw factored in 12-13 servers for 1,000 users, so it seems reasonable to assume that the power draw per desktop would be fairly low as well.

Here's the bottom line. Organizations need to do their own calculus as to whether this makes sense for them, but they need to neither drink from the VMWare ROI calculator kool-Aide, nor be dismissive of the possibility of savings. Take into account what you pay for power in YOUR area, the type of servers that you use, the power draw of the end user devices -- indeed, the systems management burden on and associated staffing of IT plays a major roe.

Even the readers who were enthusiastic about their VDI deployments were cautious to say, that "it may not be appropriate for all of your users," even though they felt it was highly appropriate for most of their users.

So - do your homework, put together a quick and dirty spreadsheet, and let the numbers talk to you. I'd be interested in hearing your stories.

Jonathan Feldman is an InformationWeek Analytics contributor who works with IT governance in North Carolina. Comment here, write to him atjf@feldman.org, or on Twitter at @_jfeldman.