A recent announcement from IBM has thrust the notion of “hybrid” into the business-technology mainstream in a big way: the company has placed a US $34 billion bet on a hybrid-cloud strategy. IBM is acquiring open source, cloud software business Red Hat.
If a prominent industry leader is considering something to be worth billions, it ought to be worthwhile to reflect on it, even if it is only for a few minutes. Big things often contain, within them, some ideas that are broader and foundational — ideas that have applicability beyond the domain where they happen to be explicitly called out.
The key idea that we will zoom into, in this Advisor, is the notion of “hybrid.” We will explore the nuances embedded in the concept, and see how the notion can help us think about architecture at a broader level.
What is “Hybrid”?
Sometimes words get in the way by obfuscating the reality underneath, but sometimes a term enters the mainstream because there is some organic meaning to it, although it may be mixed in with marketing, poor abstractions, and other sources of confusion. Therefore, it will be helpful to get at least a baseline definition of “hybrid” in place before we build further upon it.
Implicit in the notion of “hybrid” is plurality; without more than one thing in play, no hybridization would be possible. Also, we would not be speaking about “hybrid” if there were independent and isolated capabilities; the things that are part of a hybrid must tie to one another in some way. There must also be some similarity of purpose among the parts of a hybrid system; otherwise, they would just be complementary components in the system that they are a part of. At the same time, they must also have important differences so that they are not just different instances of the same general class, as object-oriented people might say of such a system.
So, we can say that a hybrid has more than one thing, those things are tied in some way, and the components have some similarities, although they are different in some meaningful ways.
Having established a baseline definition, let us loop back to the example that inspired this exploration of the notion of “hybrid.” In a “hybrid cloud,” we have two kinds of clouds: public and private. They are not just two complementary components in a larger system; arguably they are each, in their own way, different implementations of a similar capability. The two have similarities in that they both bring similar abilities to abstract away physical infrastructure and operational concerns, while being different in the variety of services, internal integrations, and controls that can be exercised by the enterprises that choose one or the other. Or, both. That last part — almost an aside — is particularly important.
One is Greater than Two!
There is something magical that happens when you bring two things that are similar, yet different, together. They start to come together in a way that makes it harder, over time, to tell where one begins, and the other ends. They may retain their distinct identities, however. It just depends on how you look — as with the optical illusion that seems to be two human faces staring into each other, but can be teased out into the outline of a wine goblet if you roll your eyes in a slightly different way.
There is an emergence that occurs, as a broader notion — that of a “platform” — comes into view to enable the individual parts to be considered as a unified whole and acted upon as one thing rather than two. In our hybrid-cloud example, it may be desirable to, for instance, be able to deploy applications across the emerging cloud platform; some parts of the application cast in one mold, while other elements embrace the yang to the former’s yin. And, allow the application to remain whole in conception to the human eye, though distributed in reality. That is important, too. The human eye — more correctly, the eye and the brain, together — deal better with fewer rather than more things. The emergence of one from two may well be driven by that subconscious tendency toward simplification.
Simplification is almost an imperative heuristic when we have a need to deal with things unseen but which will be in our futures. Enterprises and organisms, almost by definition, have futures.
A Hybrid is a Skybridge to the Future
As we get familiar with one way of doing things, our thoughts extend beyond to fill the gaps in it that we become painfully aware of, including those from demands that emerge as the future unfolds. Newer ways of doing things are conceptualized, but the buildings of the present may pose limitations to realizing those novel ideas. We need to be able to cross over to other buildings so that the future can sit alongside the present as an equal citizen.
A hybrid architecture allows us to introduce elements of the future while respecting our investments in the past. And, while giving consideration that moving the past into the future may occur at different speeds in different parts of the enterprise.
Being hybrid is not new. Not at all. If we look around, we can see the entrenched old and the emerging new sitting side by side in our enterprises. And, in important instances, we have actually thought through, at some level, how the old and the new are tied together, thought of together, and managed together.
If we do it right, we can build hybrids that allow us to have one foot in the present and one foot in the future, as we make our way across the skybridge to the new buildings of the future.
How many hybrids and skybridges do you have? Do you know where they lead to? Post your comments at the link below, or send email to bprasad@cutter.com.
