The cost of a Virtual Desktop Infrastructure (VDI) is one of the things that will get us all to stop and think for a moment…what problem(s) are we really trying to solve?
VDI is a premium way to deliver workspaces, thus it comes at a premium cost. It’s not cheaper than a PC, but that’s potentially a good thing.
What problem(s) are you trying to solve?
VDI is deployed for many reasons; remote access, simplified image and application lifecycle management, data sovereignty and security, perhaps BYOD, contractors, or seasonal burst requirements. These are, however, the problems we are trying to solve, NOT the justification for VDI.
If there is one thing that the pain and suffering of getting VDI right has taught us over the years – is that it’s hard to get it right. Embarking on a VDI deployment should come with the warning: please don’t run with scissors.
There is a long line of vendors who have made gradual advances in the right direction, with even Microsoft recently providing licensing cost relief. Along with the new hyper-converged and storage offerings available today make VDI of the 2008 era look positively archaic.
But…are we all standing on a whale fishing for minnows?
When comparing the cost per workspace of VDI and claiming it is cheaper than a PC we are doing a disservice. It’s like trying to compare Apples to pickup trucks. If cost per workspace is your goal – VDI isn’t going to work for you. Trust me on that.
My experience of VDI, along with the deployment process of pilot, proof of concept, competitive analysis, deployment, scale-out, etc. has taught me an awful lot. I’ve learnt that data, profiles, applications, and visibility not statically bound to a DEVICE is several orders of magnitude better than dependence on x86 hardware and monolithic stacks.
Data, profiles, and applications are services we use when we need them – not fallow objects of inventory, and certainly not objects to be locked in databases. IT should look at the value of abstraction, separation, and assembly upon use.
So, I submit to you the greatest lesson I have learned in the past 8 years of the desktop; if you get your users under management FIRST using data, profile, and application management and layering, THEN choose your Windows delivery methods and devices (physical PCs, VDI sessions, XenApp, RDSH, Cloud, tablets, smart phones, BYOD, etc.) and LASTLY choose your vendor(s). Not only will you save money, you will not fall victim to believing VDI solves everything.
VDI should stand for “Very Distracting Idea” – because it is but one course of treatment, it is not the cure. Delivering network stored data, profiles, and applications to a physical PC might be the cheapest, but not necessarily the best. And when I say best, I mean from all measures of productivity, not just CAPEX. This is inappropriate maths, just as it is to compare the cost of a typewriter to a PC.
The new maths is a calculation of the productivity gains to three stakeholder groups. 1. Users 2. Budgets 3. Staff. sIf the net result is an increase yield in productivity – the technology you are exploring is justified. If not, please rethink your decisions. And if you cannot measure it, you cannot manage it – don’t overlook monitoring, these distributed workspaces have a few moving parts and proactive visibility is key.
You cannot view VDI as a cheaper alternative to a physical PC environment, but VDI can deliver productivity benefits beyond PCs and this is a good thing!