The stated goal is usually around saving money, although there are certainly other benefits.
Newton's Third Law states that for every action in nature, there is an equal opposing reaction, and consolidation is no exception.
I think IT vendors tend to push consolidation without a full appreciation for the very valid opposing forces. And I think IT organizations sometimes lack the will to overcome these resisting forces in pursuit of a better outcome.
Case in point: now that I'm at Oracle, I see these amazing cost-saving proposals around database consolidation. We're talking eye-popping big savings numbers, buttressed with a rock-solid justification. But I don't see nearly enough leaders acting on these opportunities.
Nobody's right, nobody's wrong -- but it shouldn't be that way.
The Simple Ideas
You could virtualize them on 20 individual servers, or run them all in a shared cluster of maybe 10 servers.
Your cost-to-serve drops dramatically as you push the consolidation densities.
Take 100 storage footprints. The same discussion applies: you could store them on 100 individual devices, do a bit of sharing across 10 arrays, or buy one big honking storage array for them all.
As consolidation density goes up, efficiency can increase in multiple dimensions: dynamically shared resources are more efficient that static reservations, operational effort covers a larger domain, certain features and services are more cost-effective to apply, and so on.
Consolidation up, cost-to-serve down.
It's pretty compelling stuff. But it's not that easy, is it?
The IT Vendor Angle
Every year, IT technology gets faster and cheaper. More performance, more capacity, more features -- and less $$$ per unit.
But not all workloads are growing as fast as the technologies that support them. This leaves IT vendors with a challenge: how do you justify the need for the faster, bigger and more efficient thing?
It is not all that difficult for a vendor to come in, look at your environment, and see all sorts of opportunities for much better results through consolidating what's already there.
So they put proposals together that show compelling impact, and then wonder why more IT leaders don't immediately act on them.
Because people, that's why.
The IT User Angle
If you've spent any time in real-world IT organizations, you'll have an appreciation that consolidation can have all sorts of nasty sharp edges.
Right off the bat, most end users of IT like having "their" environment: their hardware, their versions of software, and their designated people to run it. After all, they're paying for it, yes? Call them server huggers if you like, it is what it is.
To consolidate 100 workloads, you may have to convince 100 "owners" that it is now in their best interests to throw their lot in with 99 other folks. Hint: they won't be swayed by your cost-saving and efficiency arguments.
Remember, these same people probably drive to work when public transit would suffice, like having their own offices, etc.
They will all have to migrate to whatever standard you've chosen for your consolidation tech. Balancing resource usage becomes important. Upgrades and migrations used to be minor affairs, now they're not. And heaven help when a problem knocks all 100 off the air at once.
For every compelling consolidation proposal, there seems to be an equal and opposing force.
Ways To Move Ahead
Not every workload deserves to be consolidated, and not every workload can afford to remain unconsolidated. How best to strike the balance?
The trick, I've found, is to expose options (and implied costs) to IT users.
The worst I've seen is "flat tax" IT funding models. This is where every business unit gets levied an "IT tax", with little regard to IT resources actually consumed. Light users grumble about the unfairness of it all, and heavy users invest serious time and effort to get more.
Normal human behavior, really.
The best I've seen is when IT creates choices and exposes service costs back to IT consumers. Here is our shared (consolidated) service. Here is our dedicated hosting service. Go ahead and choose, we'll just make sure that when it comes to budget time your choices are duly noted.
You don't need to be in a full chargeback model to see this work effectively. Even a modest degree of showback is enough for business users to make reasonable choices.
People will drive all over town to save two cents a gallon on gas, right?
Imagine if I had to run my business unit with a "flat tax" for labor -- I got charged a fixed sum no matter who I hired, or how many?
Of course -- we all get to see headcount costs, as that's a critical input to any business model.
Well, folks, so are IT services. It's 2015.
I clearly remember the day many years ago when I got my first "bill" for IT services consumed by my group. Let's just say it led to a lot of interesting discussions. Prior to that, it was all just a magic black box in the sky. My all-you-could-eat IT cloud.
Once I had a bit of detail on what my choices actually cost, (and picked myself off the floor!) it changed my perceptions and behaviors.
And I've seen this effect play out literally hundreds of times. If you can't hand your IT users an actual bill for services provided, then make up a good fake.
We're all used to getting a bill for what we consume.
Give people an option to control their IT expenses, and you'll find far more willing and eager participants in your next consolidation project.
Consolidation Is Inevitable, Or Should Be
Workloads and data continue to proliferate -- it's part of thriving in a digital economy.
Yes, the supporting technologies are always getting faster/cheaper/better, but the operational model has to change to take full advantage of what the new tech can do.
Call it IT-as-a-service, call it a private cloud, call it whatever you like -- it doesn't matter. They're all fancy names for the same idea -- consolidation of IT services and resources.
Yes, expect resistance and challenges. That's inevitable -- it's all part of leading an IT organization forward.
Think of it this way: the sooner we get comfortable in doing much more of this, the sooner we'll have more resources to invest in the innovative side of IT.
Like this post? Why not subscribe by email?