Published January 2016.
“[…] the operating model of many IT departments is shifting away from capital expenses (CapEx) when possible and toward a primarily operational expense-funded (OpEx-funded) model for completing projects. A large component of recent successes in these areas is due to a shift of non-premises, corporately managed resources to public cloud infrastructure and Software-as-a-Service (SaaS) platforms. Since cloud resources can be billed just like a monthly phone bill, shifting IT resources to the cloud also shifts the way the budget for those resources is allocated.”
In “Building a Modern Data Center – Principles and Strategies of Design”
All of technology is a manifestation of yearning. The dominions of fabled gods and heroes are the real drivers of invention. Well before SF mythology planted the seeds. But I’ve yet to read a SF book wherein the Cloud Computing concept and technology is accurately depicted. While I was reading this book, I kept thinking on Alan Turing, who wanted to create a machine that could think and be smart; but remember that he thought of that in a world where there was none of that, i.e., he created machines and started that road of newness. In this day and age, it’s very easy to think things up, because we’ve all sorts of technological stuff and the only thing that I find closer to what Turing was looking for, are the algorithms that detect my pattern of preferences in a web browser and make suggestions to what I should be buying. We are so far in having that AI that is able to pass the "Turing Test". I wonder if he were to be alive in this era, how disappointed he would be, the machines that can think only exist in SF. The SDDC is one of those not so far-fetched concepts taking place as I’m writing this. 10 years ago I can guarantee it was still in the realm of SF… Who would have thought in 10 years we’d be talking about concepts so technological advanced?
I’ve been working in and around Data Centres, in several roles, almost as long as my career in IT. At the beginning of my IT career, more than 20 years ago, every time I had to deal with the Data Centre I used to start fuming. Everything was a bitch to implement. One has to take into account the concept of Cloud Computing hadn’t been invented yet. There were no SDN, SDS, SDC, HCI, and SDDC acronyms in place…SDx had yet to come into the fore. When I started working more closely with the guys coming from the depths of IT, I felt certain that deep in the bowels of that realm there were bearded geeks in flip-flops. Nowadays the bearded guys are replaced by pimpled youths in Adidas, and it’s those who are dreaming up more SDx three-letter words. Either that, or the people from IT are being told by sales: “Hey, we need more three-letter word technologies; can you guys come up with something for us to use and earn a few more euros?” As I’m writing this post, the SDDC is just a bunch of virtualised servers, (in some cases) of software-defined storage, and SDN. I’m not sure the SDDC technology is there yet. In my book, SDDC only can be substantiated by the full virtualisation of the three cornerstones of data infrastructure: SDC (Software Defined Compute: Memory and CPUs), SDS (Software Defined Storage - the place where we put our stuff, allowing for deduplication, compression, pooling of different types of storage media, and so on), and SDN (Software Defined Network - the stuff that gets things from one place to another). With the possible exception of the SDN, none of these technologies are at a point where they can be put together into one seamless, and trouble-free computing ecosystem. I know what you’re thinking as you read this lines: “But Amazon, Google, Facebook, Google, and Microsoft already use infant SDDCs”. Quite true. But don’t forget any one of those companies have the staff to work the kinks out of them. Without the proper staff forget it!
Lowe’s book has some good stuff by incorporating research data from more than 1,200 IT pros from 53 countries, also tracing the evolution of enterprise data centres, from legacy islands of data through virtualisation and cloud computing, to the rise of software-defined centres. As stated in the book, the question is not whether we should adopt cloud computing or not. The crux of the matter is whether we’ve found the right implementation to go along with our own business needs. Nowadays we all use smartphones, web email, have Facebook accounts, Google accounts. Why do you think this is happening? Are we all stupid? Of course not. Cloud computing resources are omnipresent and it’s the new way we interact and exchange information with each other. There's no way around it. It’s more efficient, more social, more scalable and I dare say more secure. It may also lower cost depending on the type of implementation. This benefits everyone. Each business is the way in into this ecosystem, making them social and giving them a voice. The only thing Lowe, Davis and Green don’t say is that Cloud Adoption is full of deadly perils…The picture they paint seems too rosy for me, and I’m in the trenches, so I know what I’m talking about. I’d have liked to see more stuff on the pitfalls-side of things. There are a lot of them. The IT Data Centre guys doing the implementations have to be aware the road is not paved in gold all the way; there are also a lot of mud and pot-holes…they must bring a dinghy and their Wellington boots along for the ride when participating, running, or sponsoring an infrastructure project aimed at taking my beloved stuff into the Cloud. Mum’s the word.
SF = Speculative Fiction.