It took me less than 24 hours of being back in my former “home” of the Middle East (Abu Dhabi to be precise) to be starkly reminded that I had been remiss in my goal of penning some thoughts on the potential emergence of what I’ve dreadfully-named Functional Clouds and, more specifically, where I see they could play a part in the medium and long term future of many enterprises looking to gain more efficiencies from the increasingly buoyant cloud computing market.
It may come as no great surprise, given my background, that the above photograph, taken on a leisurely stroll around the immediate vicinity of my hotel, sparked me into action. Call it professional curiosity, I suppose, but this view has been representative of pretty much all of the United Arab Emirates over the last few years, but beneath the almost phototropic reach of this new facade lie some pretty interesting challenges of engineering, collaboration and construction – all of which, you can pretty much guarantee, are not done simultaneously, nor completed in the confines of the zip code where the foundations of this new development lay.
Dwelling upon another acerbic, yet clinically sobering latest thought from fellow Cloudbagger and all-round Paleo monkey, Chris Hoff, and my recent (cough, cough) “revelation” from the opening presentation at last week’s CloudCamp North (yeah, the one where I state the blindingly obvious point that Cloud is a journey. A new approach and a conduit for business transformation) I began to wonder if the basic premise of Functional Clouds might actually make a lot more sense than I had previously given the notion credit for.
First, let me tell you what a Functional Cloud is not. It is categorically not something that is directly related to the Big Data Bandwagon currently rolling it’s merry way through Cloudzville. To save you, dear reader, from abject confusion, I’ll save the reasons why it’s not directly, but is indirectly, related to Big Data for a later post, but suffice to say, Functional Clouds are a part of the mechanism that will potentially generate and localize (dare I say data gravity at this point) the raw materials of a future Big Data strategy.
Second, let’s just examine the key word. Function. Noun. Or simply – the kind of action or activity proper to a person, thing, or institution; the purpose for which something is designed or exists; role.
Now, switch your mind back to the photo and let’s take the Engineering & Construction analogy. During the end-to-end execution of any large construction project, irrespective of the ultimate facility being built, there are large teams of often geographically dispersed experts working collaboratively on the Engineering, Procurement & Construction components (oddly enough, each of these E-P-C components is what we in the trade call a Function) and behind the scenes, there are small armies of folks working on other critical support pieces such as Safety, Scheduling, Cost Control, Quality Assurance, etc. etc. and again, each of these is known colloquially as a Function.
It is relatively common for multiple parties to be involved at each stage, each with very specific deliverables and each with their own physical locations, processes, systems and software packages in place to fulfill their contractual requirement (deliverables). These can be anything from drawings to technical specifications to complex 3D models that are driven by quantities (commodities such as steel, concrete) and affected by schedules. There’s data and there is lot’s of it. Even more interesting, the stages and sub-stages of a project have tons of interdependencies and interoperability needs, like a huge supply chain, all coming together via one master schedule that has a component that requires a razor sharp focus to ensure everything stays on track. We call this the critical path and ideally, you never want to be on it.
Speaking of schedules, typically, all construction projects, by their very definition, are finite. Durations can and do change, depending upon scope, but the one thing that all projects have in common is an end. A finished facility. A drop dead date. End of. Nada mas. Nient’ altro. Khallas. However, don’t confuse this with the operations and lifecycle of the post-project facility when the engineers and constuctors have turned the proverbial lights off and when the gazillions of documents created during the preceding phases take on an extra significance to the ultimate owner / operator. (See, I told you there’s a separate blog post that will explain the rich pickings of that Big Data story).
And breathe. In and out. Isn’t this a lovely blog post? But what has it got to do with Cloud and especially Functional Clouds?
Let’s examine what we learned so far. Global engineering and construction requires global collaboration on a massive scale. Each Function can be distributed across geographies and companies. Speed is critical to success. Functions are responsible for the creation of deliverables and the timely execution of their work. Each project is a temporary enterprise. Huge amounts of data are created, stored and exchanged during the life of a project and beyond. Interoperability is a key tenet. Sound like a compelling case for many of the “selling points” of Cloud? I think so.
The applicability of Functional Clouds goes far beyond the concrete and steel world I live in, across manufacturing, financial, supply chain, even to pharmaceutical R&D – anywhere that the challenges of the above paragraph present themselves. I think of this concept as almost the next generation of today’s SaaS model, where the acqusition, coupling, flexibility and global availability of fit for a given function services is offered via industry marketplaces to the next generation of CIOs – who I have long posited will become service aggregators to their business.
Are we there yet? No. Not by a long way. But the shoots of this are certainly starting to sprout. There’s material evidence that I’m not completely bonkers, which I find somewhat reassuring. I suppose one could look back on the work that was done by Citrix, Dassault & Boeing in 2007 during the design of the 787 Dreamliner as a significant waypoint along this path, and more recently, the notion of “special purpose clouds” was discussed on Focus by some notable contributors almost a year ago.
Finally, we’re seeing evidence of plays in this space by Autodesk, IBM & Fujitsu and reading between the lines of Chris Hoff’s comment – I wonder if the smart CIO will be the one that ignores all the vendor hype around “how to do your cloud” and becomes the one that focuses on the tale of “how someone else did it for me”.
(Cross-posted @ The Loose Couple's Blog)