Architecture of On-Premise Systems

Do you think all of the “aaS” elements change the architecture of solutions built for on-premise systems too?

I’ve been reading and hearing more and more that the on-premise architecture bits are expected to take into account cloud bits and pieces too. With things like “database stretch” with SQL Server and all of the functions and services and so-on that you can incorporate, at the very least, it adds questions to the mix.

The questions have to do with things like:

– does it have the security I need?
– how’s the performance?
– how does it integrate with my on-premise systems?
– what can be done better (and what’s better about it) with these options?

These last two elements are key. In fact, the integration question is something that seems to be overlooked too frequently. It seems like the questions revolve around the “cool” feature or the hot feature of the day, but the integration ability comes later.

I think we’re moving more to seamless integration. You can see this with the database stretch option specifically, it just appears and steps in when needed. I think we’ll see that more and more as pieces of the application puzzle come from services – and then options and choices will have to come from where those services may originate. It could well be that you host your own functions as a service, for example. But I do think it will be seamless and that we’ll all hopefully be picking best fit for a situation or challenge, rather than having to necessarily look at where the computing power is coming from to fulfill that.

I do think it’s a bit like self-driving cars, I think it’s a (relatively) foregone conclusion that we’re headed that way (and I say “Bring it on!”) but getting there from here will have successively strong and well managed steps along the way.

There’s a good bit of caution to apply, of course. One of the things that I’ve found a bit worrisome is the reliance on some of these services only to see them cease to be developed further or worse, discontinued. I don’t know how to deal with that, really. If you’ve built solutions on top of some feature that stagnates or worse, you’re in a much worse spot than you were before you chose that direction.

While I know that could happen with literally any software, solution, tool library or other component we collectively pick, many of these “aaS” items become so key to the solution.

We just have to manage this process to get there; and choose wisely along the way and have “Plan B” type things thought of I think.