Seeing is Believing: Ignorance of What Goes Over Your WAN is Not Bliss

“You can’t manage what you can’t measure,” is cliché in IT executive and management circles and has been for decades. It has driven the deployment of myriad management tools (and birthed an industry) and fueled the ongoing rise of IT Service Management.

The necessary corollary is, “You can’t measure what you can’t see.”  Without instrumentation to provide visibility into whatever part of the infrastructure we are managing, we can only manage based on guesses, good intentions, and user complaints.

And yet, we have for more than 15 years spent a huge chunk of our IT budgets on WAN services with only the scantest knowledge of what is going on across them.   We can easily see low level facts such as latency, loss, jitter, and utilization…can we easily see how much bandwidth each application and each user are getting? How often the performance of a key application fell below our SLA? How many times carrier reliability problems affected peak production workloads? And, despite the huge blank spot on our WAN radar screens, our enterprises are flying further and faster into new, WAN-­‐dependent territory. WANs continue to get more important to the ability of the business to do business and even to find and get new business.

Certainly, we are not totally blind.   Routers, firewalls, and other devices can provide some information about link health and utilization, for example. Unfortunately, they can’t usually provide the kind of detail a business serious about investing in IT to enable and drive business wants.   If we want to know how much WAN capacity is supporting specific sales-­‐ and client-­‐services-­‐related applications, and how well those applications are performing, the router is not going to tell us what we need to know. WAN optimizers and traffic shapers can provide a lot of application performance detail, and can usually even control priority and performance. But, they are rarely deployed everywhere due to cost, and so cannot supply the info we want across a whole WAN.

The point remains: Seeing how people actually use the WAN is the first step toward taking proactive control of it and reshaping it to better serve the organization’s needs. So, IT needs both broader and deeper visibility. In evaluating any router update, any managed WAN service, any SD-WAN solution or NaaS offering, IT should be digging into (and wherever possible test driving) the visibility features of the new solution. Ideally, their new-­‐model WAN will tell them

  • Which applications are present on the WAN (generally, and site by site as needed) – with the expectation that at first, they are going to get a lot of surprises as to what is actually going on
  • How much bandwidth they consume (and ditto on the surprises)
  • What performance needs they have (e.g. regarding loss and latency)
  • What kind of performance they are getting
  • What kind of connectivity they are using—MPLS, 4G, Carrier Ethernet, Internet, etc.
  • How much each application’s use of those WAN links costs

When IT can get that kind of information from its WAN, it is finally able to truly align spending on WAN services with business goals and outcomes: to manage the WAN as a resource for the business and invest in it appropriately thanks to full visibility into what the WAN is doing (and how well it is doing it).

To learn more, join the December 15th webinar, Demystifying the Software Defined WAN. Register here.

Categories: Software Defined WAN (SD-WAN)