Verizon's Hakl on SD-WAN Evolution

SD-WAN may be cool new technology, but for a self-confessed SD-WAN enthusiast like Verizon's Shawn Hakl, the use cases, cost propositions and actual outcomes must be made clear.

Previously, in Part I of this Q&A (lightly edited for length and clarify), Hakl, Verizon's vice president of business networks and security solutions, addressed Verizon's approach to standards and organizational shifts as related to SD-WAN from his perspective as a vocal SD-WAN evangelist. (See Verizon's Hakl: SD-WAN Delivers a Multitude of Benefits.) In Part II, Hakl talks about what the future holds for SD-WAN disruption.

Telco Transformation: What has been your learning experience in terms of how you explain and evangelize SD-WAN and the rest of Verizon's SDN strategy to customers?

Shawn Hakl: What I would say is that, with SD-WAN and the virtualized NFV stuff that we've done, we have to be even more conscious of the whole "people don't buy new technology for the sake of technology." The problem is people can look at the stuff as just pure infrastructure, in which case they have very little interest in understanding the technology. I haven't focused as much on educating on technology as we have on demonstrating the use-case benefits. With the telco market, it's become fairly entrenched. You rarely have to explain to somebody what a router is or how it works, right? But when you're talking about the effect of deployment of SD-WAN in combination with, say, for example, virtual firewall, you're re-architecting the way you deliver the service if you want to be efficient with the technology, and that is a different skillset.

People had a utilitarian version of how to select telco infrastructure. And because there are so many disruptive choices out there -- you've got new vendor, you've got very different choices for implementation in terms of how people approach this technology that you have and recognize the benefit of technology -- you have to spend more time on it. If you're going to make that connection, you need to understand why from a business perspective. And so working with people to understand how the impact of virtualization affects the network translates into a reason to expend the network.

Once we can draw the analogy, and I don't think it's too big a stretch for people to understand, you can see what happens when you're talking to the CIO. Every CIO knows what happened when virtualization hits the data center. There are a couple of ways that it can instill a tremendous amount of change in the way you both operate as an organization as well as how you set up your infrastructure. And to draw the analogy that says that that same technology is coming to the network at least sets the stage that says, "Hey, this is something that will be significant and is worth the effort because it has big payoffs." But they understand, potentially, what has to go there.

Then the next stage is to take them into fairly precise use cases around where you can leverage that technology today and drive more agile operation, better customer service, security policies and better networks so you've got better protection. There are some very clear use cases, and it's helping people see that. And it's also training our own organization to move away from the traditional pitch around network equipment and make sure that people understand that this is very much an outcomes-based sale -- an outcome-based use case -- that you have to be able to demonstrate.

---

Want to know more about the companies, people and organizations driving developments in the virtualization sector? Check out Virtuapedia, the most comprehensive online resource covering the virtualization industry.

---

TT: Am I correct in saying that SD-WAN leads into SONs [self-organized networks] a little bit? Or how do you see this centralization of infrastructure management in the virtualized setting ultimately culminating?

SH: Yeah. I absolutely see them as related. And so, whether it meets the precise definition of SON -- apologies, just because, often, terms like that come with a lot of connotations -- I will tell you that, yeah, this is our view of how you achieve a fully autonomics-based network. So I want a network that can self-heal, and I'll use an example of that. If I have a virtual process that's connected, that I know how I installed that network and which physical elements I've got, then if it starts to degrade, there's no nominal cost for me to restart that. So I can watch that restart and create resiliency at one point.

Similarly, if I have an organizing principle around that, I can start to move that different functionality to different hosting nodes on my network, giving it geographic and high-availability capabilities. That, from my perspective, is how you self-heal.

From a defense perspective, you can have a network that understands how a process operates within the system, and if it starts behaving anomalously or differently, you can isolate and sandbox that process, restart it, have it correct itself and only if it continues to exhibit bizarre behavior do you have to actually intervene. Otherwise, you can sort of self-defend it with that.

And then, finally, if you have a million patterns of data that have optimization algorithms that you can identify, you can insert optimization algorithms to speed up that data transmission. You can start to look for similar patterns that you don't know are the same data, but you can guess, try it out. If it works, great, promote that optimization into production, and if it doesn't work, then you haven't lost much; leave it out of the network.

So in terms of the self-organizing, self-optimizing, self-defending network, that's how we see this transpiring. And then this other notion that you can have pieces of the software in the network broadcast its ability, and then start to build policy by intersecting the data path you want with the policy that's available to various nodes in the path to match your policy, then that's absolutely where we see this stuff going. And some of that we've got done in some way, shape or form already, and, of course, you'll keep evolving and expanding those capabilities over time.

TT: Do you have any final thoughts related to SD-WAN and this next generation of solutions?

SH: This is not a technology change. This disruption doesn't come from the coolness of the technology; it comes from the neat things that it enables. There are very few organizations that would turn down an opportunity to be more confidently secure in the use of the resources in the company. There are very few organizations that would turn down a chance to better manage costs. As we move into that sort of digital and solutions design for the digital native I think, and I'm stealing some marketing talk, people are looking for solutions to scale their business as they try to digitize their entire business, and this is the technology that enables that. So I think you'll see a lot of people interested in it as they run into these problems within their business and they're looking for technology solutions to solve that; these are the places they're going to turn. The level of innovation that's going to come out of it, we're just at the tip of it and it's going to be exciting.

— Joe Stanganelli, Contributing Writer, Telco Transformation