Data Packets Don’t Lie: Exploring the Promises of SD-WAN

In this follow-up piece to my training session at ChannelCon, I’m going to lay out the various angles that service providers attempt to use as wedges with their customer base as well as the questions that end users need to respond with when they hear this type of rhetoric.

David Landsberger speaking at ChannelCon 2018I spend more time training the channel on software-defined wide area network (SD-WAN) technologies than any other type of service or technology, and over the past few months, I’ve noticed a plethora of service providers fast-following with SD-WAN or software-defined networking (SDN) technologies. Each one of these service providers has their own marketing angle that they look to exploit when positioning with customers and IT practitioners.

As IT professionals, we need to remain vigilant. In this follow-up piece to my training session at CompTIA ChannelCon, I’m going to lay out the various angles that service providers attempt to use as wedges with their customer base as well as the questions that end users need to respond with when they hear this type of rhetoric.

Additionally, I’ll add in some further education on these talking points, so you can remain informed and empowered when evaluating these technologies The goal is to make sure we are digging in as deep as possible and measuring what these face-value statements mean, all while building our base knowledge of SD-WAN.

3 Claims You’ll Hear About SD-WAN and How to Respond

1. “We built (insert SD-WAN provider) into our core.”

Service providers like to wear this one like a badge of honor, but this statement doesn’t really mean anything. All it means is that the service provider is hosting the SD-WAN application within their own data center network.

I suppose there’s one less hop for the transfer, but that will not affect the clear majority of customers. Or, the service provider could have possibly built a few application programming interfaces (APIs) or integrations to enhance the analytics of the SD-WAN portal on the back end, but most are simply offering a re-skinned version of the graphical user interface (GUI) that the SD-WAN provider already had.

The reason why this doesn’t mean much is that we are talking about SD-WAN and not SDN. SD-WAN allows a customer’s network (key word: customer) to decide with what priority and resilience each of their outgoing data packets receives when leaving the corporate network. To use a metaphor, it’s like deciding which car gets to leave the driveway first.

SDN refers to the actual core network itself. This means the entire network infrastructure of the given service provider is routing traffic with software-defined priority.

Why does this matter?

Well to extend our analogy, you can decide which car leaves your driveway first, but once you hit the public road and are in the middle of multiple lanes of traffic, can you control which car goes first? No, you can’t. There are too many variables. SDN takes things one step further by promising that you would be able to prioritize packet transfer throughout the entire length of transport.

Private transport like dedicated internet access (DIA) connections or multiprotocol label switching (MPLS) can soften the impact, but many customers are exploring SD-WAN technology to get rid of those expensive circuits.

SDN is the end game of SD-WAN, and SD-WAN is simply a band-aid until we get there.

Major telecoms across the world are in the process of rip/replace and augmenting their core to be SDN focused. But it’s not finished, and that’s why this statement is an empty one. Until most major telecoms have completed this upgrade, no data packet could possibly be transferred end-to-end with software-defined priority.

QUESTIONS TO ASK:

  • What additional analytics or services does your implementation of SD-WAN offer me versus directly procuring it from (insert SD-WAN manufacturer)?
  • Is your network fully SDN? Where is it, and where is it not?
  • How do you enforce my packet priority once my data packets leave your network?

2. “We have the best SD-WAN support.”

It’s not that service providers don’t have good support for SD-WAN; as a matter of fact, they do. It’s simply a matter of demand and scale.

I work for a national technology distributor with more than 3,000 active partners. In 2018 alone, we have seen SD-WAN jump into the top five of our most lucrative technology solutions.

In 2017, SD-WAN wasn’t even in the top 20.

How can a service provider take on a multitude of new customers and effectively train and onboard their SD-WAN support specialists? It’s a real personnel challenge. End users are having issues receiving timely support in the implementation process.

Add onto that the massive consolidation that we are seeing in this space, and you have the perfect storm for a poor customer experience.

This is why local managed service providers (MSPs) have such a tremendous opportunity with SD-WAN. MSPs can take on the management of the implementation and rollout and can toe the line on when they reach their break point in terms of service and support. Right now, the carriers that manage these services are not turning anyone with a pulse away, and it’s hard to fault them for that.

It’s important to understand that while SD-WAN may be an easier technology to implement than MPLS, it is not without its challenges. Daisy chaining your equipment to the SD-WAN appliance, scheduling network events and creating site profiles are all more task-intensive activities than they are originally positioned as. Be aware of this as you explore the technology, because most end users are not going to like waiting on hold for support for a technology they were sold as “easy touch.”

QUESTIONS TO ASK:

  • How many employees do you have specifically supporting SD-WAN? How many sites do you have on your SD-WAN service?
  • What is your average time to ticket resolution? How does that number compare to SD-WAN ticket resolution six months ago?
  • Take me through your customer triage process for SD-WAN escalations.

3. “SD-WAN will kill MPLS.”

This is the biggest lie of them all. To be honest, the only providers brazen enough to utter this sentence are those that don’t have an MPLS solution. <sniff> Ahh…can you smell what marketing has cooking?

SD-WAN does not deliver quality of service, and it does not possess a service-level agreement. That’s a showstopper for risk-averse IT shops.

The reason MPLS will not die is because customers won’t let it. Financial firms, health care systems and governments can’t afford to sign up for technologies that offer them zero recourse in case of a service interruption or worse. It’s just bad business.

While I have seen IT shops thin out their MPLS networks, I have only seen two handfuls of customers that managed to completely remove their MPLS solution, and it was usually because they didn’t need it in the first place but used the solution to solve virtual private network/virtual local area network (VPN/VLAN) issues.

QUESTIONS TO ASK:

  • Does your SD-WAN solution possess quality of service? (The answer is no. Don’t let them lie to you.)
  • If my service goes down, what are your timelines to remediation, and how will I be financially compensated for the service interruption?

Make no mistake, SD-WAN is a real technology with tremendous benefit for branch locations and low intensity environments. But it’s a unique technology in that it’s at the peak of its hype with also being at the breaking point of its implementation ability. There’s a tremendous opportunity for IT practitioners to take advantage of this environment by investing in SD-WAN skill sets or by bridging the service gaps left behind by the big players.

Be vigilant out there, folks.

Did you miss ChannelCon? You can still get continuing education units (CEUs) by watching ChannelCon online.

Email us at [email protected] for inquiries related to contributed articles, link building and other web content needs.

Read More from the CompTIA Blog

Leave a Comment