Budget?
Surely you would include the cost of consultants in the initial budget rather than decide afterwards?
So, you’ve made the business case for your desktop virtualisation project and you have the budget to do it. The next question is whether you need outside help, either from product vendors or consultancies. How do you decide? No two organisations are created equal, and neither are IT departments. An organisation’s individual …
As an IT manager who has recently taken the companies infrastructure through virtualisation, I know from experience that its very difficult to achieve everything internally, attempt to do this and you'll run into pitfalls and problems that could have been avoided if you called in outside help.
Finding the right consultants means you'll have access to somebody who knows
The best way to implement the changes
The common problems that occur and how to avoid them
Get proper training on managing the systems
The business case for VDI is anything but straighforward other than every consultant+dog looking for their next revenue stream.
hmm you have 5500 PC's with OEM XP on them, so lets invest in a San with enough IOPS to cope with servicing all those virtual desktops, x number of servers, 5500 WIndows 7 FPP licences AND your blood sucking 5500 MS VDA (pay pe rmonth forever) licences (or alternatively sell your soul+dog to MS for an SA agreement)
Then replace your core2 duo 2Gb dell pc's that cost you £320 with thin client devices priced at what the market will bear (%10 less than a PC) and not what they are worth $100 ???
Why wouldn't you do it................ :-)
Declaring an interest in this, since it's what my company does for a living, I'd say you need to be near-crackers to do this yourself.
It's not just about virtualising stuff, it's about knowing in advance what works and what shouldn't be virtualised. There are some applications that cry out for their very own fat client PC (badly behaved memory and cpu hogs) which, if they are mission-critical for the user, are going to be very bad news.
Also, a good design would split off the users who need full virtualisation from the (probable) majority who will be entirely happy with something like Terminal Services, a very much lower cost option.
I'm highly sceptical of involving just 'consultants' who don't know what they are talking about, but when you can find real subject expertise out there, it's foolish not to draw on it.
And there are plenty of solutions which don't involve replacing all the desktop real estate at once. Consider gutting your existing desktops and turning them into thin clients if the exorbitant price of bespoke thin client devices looks too high. You get the management cost benefits if not quite the power saving.
This isn't rocket science - but it's a specialist area and what's the point of discovering all the 'well known' problems for yourself?