It can be the same with IT costs. 2 out of 3 x86 workloads will be virtualized this year, which makes cents (pun intended). Organizations picture themselves going from 10% utilization to 80%, and plot how much money they’ll save as they avoid or delay hardware expenditures, even while growing customers and capacity.
“Waste not, want not” is the name of the game.
When it comes to Converged Infrastructure, expectations of expense management are higher. More automation means more revenue, faster, right? Redeployment of under-utilized resources should be a snap. A sudden business need for a new development system can be met with a mouse click, and not a meeting. Think of the time savings!
Not so fast. If you have a colleague who thinks IT staff will be sitting around, twiddling their thumbs because you’ve installed Converged Infrastructure, have them watch this video.
Remind them that “industry standard” staffing ratios from just a few years ago have gone by the wayside. We see IT Managers snicker when presented with “20 servers per FTE” numbers, because their teams run twice that, or more. And try to juggle that heavy load with a bag full of monitoring tools that don’t agree with each other (if your colleague doesn’t believe you on this, have them check out the recent Forrester survey that found 69% of organizations use 10+ monitoring tools, that work NOT AT ALL WELL together). As Kent eloquently describes, Converged Infrastructure makes this mess of tools much worse. And much more time-consuming. Which equals more cost. We squeeze the cost water balloon for utilization & agility, and a bulge appears in the weak spot over by operational efficiency.
Tell your colleague, too, that Mean Time to Recovery (MTTR) is by no means guaranteed to go down with Converged Infrastructure. You may have spare capacity AND the means to move loads around. But if you can’t find a problem quickly, costs go up. I love how Kent delineates between the “as-running” infrastructure behind an application or customer, as opposed to the “as-imagined” or “as-built” configuration. If your documentation is weeks or months behind, if someone’s made a change because they were uninformed about dependencies, etc. – you could be looking for a needle in the ether (and dreaming of the day you had a haystack). The cost water balloon bulges over in the MTTR metrics, and your organization blames IT for lost productivity, or lost revenue.
And Kent’s point about baselining on your new Converged Infrastructure is key. You have a new customer or a new application, on new infrastructure, and something looks weird. Is this the new normal? For this application? Or for this infrastructure? This reminds me a lot of when I figured out my younger son never registers fevers with those forehead-swipy thermometers. Now we know it’s his torso that heats up instead. Wish that had come with the owner’s manual. Would have saved my family a lot of urgent-care fees and sick days. Automatically introducing monitoring to gather and set baseline data right when you spin up a new IT service can save you time and money.
Luckily for us, IT doesn’t really live in a water balloon (that would truly be an unfortunate place to house your Converged Infrastructure). Don’t feel you’re cornered in an enclosed place. Because the RIGHT tool can siphon off pressure without making a mess. So you can manage IT expenses much more predictably and sustainably which is, after all, one reason your organization bought Converged Infrastructure in the first place.
Other Posts in this Series
- An Expert’s Take on Converged Infrastructure Learning Curve – Part 1: What the Business Wants
- An Expert’s Take on Converged Infrastructure Learning Curve – Part 2: The Trouble with “Agility”
- An Expert’s Take on Converged Infrastructure Learning Curve – Part 3: Two Very Naughty Words: “Chargeback” & “Showback”