If you’re an IT manager calling your internal VMware or other virtualization farm a “Private Cloud” in an attempt to prove to your leadership that “public cloud is insecure” or “I built the same thing as Amazon Web Services (AWS)”, you need to get ready for a dose of reality in the coming year.
Server-huggers beware, you might have been able to get away with it until now, but 2013 will mark a turning point in which the term Private Cloud will be permanently exposed for what it is… a capital intensive, server stacking, virtualization game.
Just because you might have flexibility to decide how much RAM you can assign to a VM, doesn’t give you the right to “cloud-wash” your internal IT operation and call it something that it’s not… because although it may be Private (can someone tell me again why it’s important to be able to touch your servers?), it’s certainly not Cloud.
Not that there’s anything wrong with that…
Not that there’s anything wrong with that private infrastructure
I’m not saying there is anything wrong with running an IT shop where you still spend lump sums of capital (CapEx) for physical resources, especially if you are working to make those resources flexible and reliable by optimizing your data center, using virtualizing, and invoking best practices like continuous monitoring and agile development.
Just don’t use the word “Cloud” because your business users and C-level leadership are getting smarter every day on the incredible economic advantages, real security story, and global scalability benefits of public cloud.
In short, selling them a story like “my private cloud is the same as AWS, but more secure because it’s on-premise” is going to begin to look childish. And worse, it will discount the credibility of the (probably pretty good and still very useful) internal IT environment that you’ve worked so hard to build.
If you physically touched it, estimated your peak demand before buying, and/or don’t have a re-occurring OpEx fee… IT’S NOT CLOUD.
The definition of “Cloud” will also further tighten in 2013, where it will be reserved only for systems that allow you to:
- transform your IT into only operational expenditures (OpEx)
- go global in minutes
- never have to guess your initial or future capacity
Despite all the marketing from old guard IT and large virtualization software companies that claim building your own Cloud is the best way to go, your Private Cloud still:
- is a large capital expense (CapEx)
- rarely allows even the largest installs to go global in minutes
- makes you commit to a upfront minimum and requires you to predict future capacity
In his recent keynote at Amazon Web Services Re:Invent conference, SVP Andy Jassy [View his Keynote on Youtube here] put it in the best perspective I’ve heard yet, giving these six simple items that differentiate the burden of private, from value of public. [Click picture to enlarge]
Andy Jassy, AWS Senior Vice President pokes fun at Private Cloud at the recent AWS Re:invent conference
It’s okay, just try a little bit… it won’t hurt you.
Remember those drug prevention classes in middle school (was it called D.A.R.E. everywhere or was that just an Ohio thing?) where the police officers would come and tell you the dangers of drugs and how they get you hooked by getting you to just try a little bit?
“Don’t even do it once,” they would say, “Because if you try it once, you’ll be hooked for life!”
Well, it seems the private cloud loving internal IT folks were all sitting in the front row during those officer presentations, because they took this advice a little too seriously and have applied it to public cloud adoption too.
The best thing about public cloud is it’s cheaper to fail than belabor conversations about whether to try it or not.
Internal IT will remain greatly relevant
Don’t worry internal IT, you’ll still be greatly needed by your company in 2013 and well beyond because there absolutely is a place for flexible, private infrastructure in today’s IT.
Organizations that have invested millions in capital on IT hardware, software, networking, and human resources would be completely insane to throw it all away today and move everything to public cloud tomorrow; however, in the same breath, I would also call these organizations insane to keep piling investment into more private resources given the extreme economic, scalability, and functionality advantages of public cloud.
Over the coming years, even very large internal IT groups, simply won’t be able to keep up with the rate of innovation, security, and scale that public cloud operations will achieve.
Internal IT will also face tough competition from rogue business users going outside of their internal IT to get what they need from public cloud with something as simple as a credit card swipe. Of course, internal IT may think the best weapon against this is a strict lock-down policy where business users get punished for going rouge; but, a moratorium on public cloud only hampers corporate innovation and creates animosity between the teams. I suggest there is another answer for internal IT… Embrace, broker, and support.
Although easier said than executed correctly, cloud brokering both public and private IT services, while supporting business users on both,will be the key function for internal IT groups staying relevant to the business and even thriving in 2013 and beyond.
More on the “why and how of cloud brokering” soon… we’ll even take a look at some tools that can (maybe) help.
Disclaimer: These predictions are based on the fact that world does not end on December 21, 2012 as the Mayan calendar predicts. If we never reach 2013, I reserve all rights to drastically modify these predictions.
Tags: AWS, Private Cloud, Public Cloud