While I'm not a personal fan of the specific term, I am most definitely enthusiastic about the ideas and concepts it embodies. Although it's a term promoted by VMware, the thinking reflects the industry at large.
Having now sat through many dozens of VMware SDDC presentation, I've realized most people prefer to describe it through a technologist's lens: control planes, data planes, key abstractions, etc.
That's sort of like describing the impact of the internet in terms of its constituent parts: TCP/IP, DNS, HTTP etc. You might miss the bigger picture of what's really going, and how it's changing all of our lives.
Sure, the various technologies involved -- individually and collectively -- are heady stuff indeed, but that's not where my real enthusiasm lies.
For me, the mind-blowing bits are the entirely new operational models that can result from the newer technology. I think that we, as technologists, have to curb our natural tendency to study and pick apart the shiny bits, and instead invest in a serious and uncomfortable discussion about the new processes and new models.
If we don't, the technology will inevitably be used to recreate exactly what we're doing today -- only a bit better.
A while back, I was driving my daughter home from some event or other, and we were hitting heavy traffic. I turned to my daughter, and asked her to call Mom and say we were going to be ten minutes late.
I fully expected her to enter a number, wait for it to be answered, and relay my message.
Instead, she told Siri to sent a text to Mom that we were going to be a bit late -- in about two seconds flat. Her approach was more efficient on multiple levels: it was faster, Mom didn't have to answer the phone, we had proof if needed that we tried to reach her, etc.
Same technology, entirely different workflow and process. I felt a bit old at that moment.
Twenty years ago, email was starting to find its way into the workplace.
I remember that the standard practice for corporate communications was to call in your admin, dictate your thoughts, the admin would type it up, you'd correct and refine it, multiple copies were made and sent in envelopes, and -- as a final step -- the document was sent as an email, usually as an attachment.
I know, that seems so quaint and antiquated these days. A positive thought: there was a lot less inbox spam in those days :)
The point is simple and often repeated -- without a heavy investment in behavioral and process change, the power of the new technology goes untapped.
How This Applies To SDDC
The potential speed and agility of SDDC is breathtaking in its potential. But unless the business learns how to exploit all this potential, it'll be as useful as driving a Veyron in downtown traffic.
IT, will -- of course -- need to learn to think differently. Acquiring a #devops mindset and capability is a minimum, but unless that's coupled with a substantial and ongoing process re-engineering investment, there won't be the all-important continual improvement capability.
My bet is that before long, these newer IT process engineers become much more valuable than today's architects and operations developers. Learn from the manufacturing industry on this one, folks -- because in the digital economy, IT is the new manufacturing.
But the business community has perhaps even more work to do. Many of them are woefully unaware of the new potential: what it's capable of, how to intelligently consume it, and how to re-engineer *their* processes around the new-found IT agility.
But it's begun.
Push Vs. Pull
I think VMware did a masterful job of broadcasting their vision of a software-defined data center, and backing that up with key investments. Certainly much more to do, but let's give credit where credit is due.
But at one level, that's a push. And pushing ideas and concepts can only get you so far. At some point, there has to be a strong pull from your customers to deliver on the vision and put it into practice.
Well, it's started. I'm meeting more customers who are buying into the vision for all the right reasons. Dig deeper, and you'll always find a common pattern.
They know that IT has to radically transform its operational model to focus on agility, speed and ease of consumption. It's hard work, but they're making progress. Better yet, they're working with parts of the business who are desperate for the results of the transformation, and are pulling IT in the right directions.
And it's this set of IT customers who are *really* interested in SDDC.
From their perspective, it's the best architectural vision that supports the world they're trying to evolve to. While prudence always dictates keeping your options open, they aren't shy about heavily committing to specific vendor technologies, including many from VMware.
I think the reason is simple: delivering pragmatic business impact matters more than potential style points on how you went about doing it.
I look for this all the time -- is IT focused on the "what" and the "why", or endlessly locked in navel-gazing debates around the "how"? It's one of the markers of a progressive business, and correspondingly progressive IT organization.
But there's another important factor in play that needs to be discussed :)
How IT Vendors Think About SDDC
If you've worked inside IT vendors for most of your career as I have, the mindset is familiar everywhere you go: what's the next big thing, what should we be investing in, and -- since resources are finite -- what should we not be investing in?
While we'll squabble noisily about the specifics in public, behind the scenes we all generally see the world the same way. The IT model is changing. "Cloud" is the proxy for the new operational and consumption models. Whatever you do, it had better work well with what your customers are investing in, and -- for time being -- in most enterprises that means a VMware-based environment, and its associated operational model.
If you -- as an IT leader -- find yourself sticking with a traditional IT model, before long you'll realize that there are fewer and fewer attractive vendor choices for your environment and the way you insist on doing things. Complicating matters, you'll find it harder and harder to find good people that want to work in the old way on older technologies.
Now, to be fair, that's still in the future -- but it's already clearly begun. For example, if you're investing heavily in a next-gen IT environment, you probably have no shortage of vendors clamoring for your attention. You just don't get the same level of enthusiasm when looking for a tape library or a COBOL compiler :) Along the same lines, the better IT people want to work on the new stuff, and do it new ways.
The early adopters of SDDC-like concepts are clearly doing so to gain a competitive advantage. The laggards will inevitably be forced to move forward due to perpetually decreasing lack of ecosystem support.
Old tech and old processes belong in a museum, and not on a data center floor.
Like this post? Why not subscribe via email?