Thrown around a lot lately, the term "software defined" has come to mean many things to many people. It's been applied to networks, data centers, WANs, and other IT areas. But to unified communications? Not so much.
Among the multiple descriptions for software-defined whatever are a couple of points of commonality for each of these scenarios. They are:
In the UC market, the lack of a software-defined model creates big headaches in large environments. Most of the UC vendors have done a great job in enabling centralized control for many operational tasks, including provisioning phone and quality of service (QoS) settings on the network. However, enterprise communications managers still have to handle hundreds of tasks in a repetitive manner, particularly when migrating from legacy systems.
When it comes to migrating from an existing system, legacy or otherwise, to a new UC solution, the process can be overwhelming because of the amount of data involved. Users with multiple devices require multiple configuration settings that cut across voice, video, messaging, presence, chat, mobility and other systems. I've talked to engineers who have done large-scale migrations that end up entailing well over 200 configuration settings per user with data spread across a dozen or so systems. Now add in multivendor environments and cloud UC, and you can readily see how UC management can get exponentially more complex.
If all of these functions could be abstracted up from the underlying core systems, then the provisioning and orchestration of the features could be fully automated – the very definition of "software defined." Ideally, the platform would take in business inputs as a way to build the policies that are then used to configure the system.
This is one of the core value propositions of business communications operations management (BCOM) products (read related No Jitter article, "Why UC Needs Improved Operations Management"). VOSS, as profiled in a recent UCStrategies post, is one of the few BCOM vendors that has embraced automation and this vision of software-defined UC (although it doesn't use that specific term).
In the UCStrategies post, Mike Frayne, CEO of VOSS, is quoted saying the company's goal is to achieve 100% automation since any level of manual configuration slows down the entire process. And, in commenting on my June 10 No Jitter post, "Digital Transformation Requires Shift in Collaboration," Chris May, VP of business development and product marketing at VOSS, made a good point on how the mix of mobile, cloud, and on-premises UC could be highly problematic for the IT team that has to manage this environment. He wrote: "Surely this is anarchy.... Shadow IT at its worst?"
I believe the industry sits on the precipice of this anarchy becoming a reality. Digital transformation is making UC more important, and UC is evolving down many paths at once. If we don't quickly embrace the software-defined UC concept, the industry risks collapsing under its own weight because the systems can't be managed without an army of people.
"Software defined" has had a profound impact on the entire network industry, except for UC. Isn't it time UC caught up?
Follow Zeus Kerravala on Twitter and Google+!
@zkerravala
Zeus Kerravala on Google+
