This website requires Javascript for some parts to function propertly. Your experience may vary.

Network (In)Visibility Is the Achilles Heel of SD-WAN | Cisco Investments

We use cookies to improve your site experience and deliver personalized content. By continuing to use this site, you consent to our use of cookies.

Network (In)Visibility Is the Achilles Heel of SD-WAN

Brooks Borcherding, LiveAction's avatar

Brooks Borcherding, LiveAction

When I worked at Cisco from 2005 to 2009, I was exposed to the early concepts of pooling compute resources and the emergence of what became cloud computing. I left Cisco to lead a company that built one of the very first Cisco based public cloud platforms. The cloud held the promise of significant growth and opportunity for those companies that knew how to take advantage of its elasticity, scalability and cost savings and we’ve witnessed the explosive era of growth over the past decade. Yet for all the benefits of cloud, it also created significant complexity for the enterprise network as workloads are distributed over public, private and hybrid cloud services. And this complexity presented enormous challenges for legacy network performance monitoring tools. Now, with the emergence of SD-WAN, the network management challenges become more complex as SD-WAN adoption continues to skyrocket. ICYMI, the worldwide SD-WAN infrastructure market is now expected to reach $8.05 billion in 2021, according to industry analysts at IDC.

While nobody will disagree that software-defined is the future for network fabrics, the problem is this: Legacy network performance monitoring and diagnostics (NPMD) tools were originally designed to manage traditional data center – campus – branch network architectures. As software-defined fabrics proliferate, a new level of network complexity will push the limits of many legacy NPMD tools. Lack of visibility into performance between sites, devices, users, and applications is chief among them.

Sure, there are lots of ways to know what’s going on in parts of the network or see the performance of a single application. Yet the ability to visualize the entire network, regardless of the origin of the software or hardware that’s on it, and use that information to make faster, smarter business decisions remains elusive for a lot of businesses.

This explains the emergence of next-generation NPMD platforms.

NPMD is no longer only about managing and monitoring the performance of certain aspects of the network or troubleshooting issues or thwarting nefarious characters. Today, NPMD is about simplifying the network so that it delivers cost savings, network agility for the line of business, and faster mean time to repair (MTTR). It comes down to having complete visibility into the network and being able to apply advanced analytics to that big picture view so you can take action before it impacts the business.

It’s not surprising that the SD-WAN and NPMD markets are concurrently realizing phenomenal growth. Arguably, SD-WAN architectures drove the need for more sophisticated NPMD platforms. In fact, current research on the NPMD market estimates it at $2.1B and growing at a CAGR 15.9%, according to Gartner.[1]

Yet getting that unified view of network performance still presents a challenge. Think about this – if you can’t manage what you can’t see, how do you know you have complete visibility?

Odds are that your network infrastructure exists as a multi-vendor, multi-platform, and multi-cloud norm. It also likely utilizes WAN transport of MPLS and public internet to better manage costs and ensure service assurance. Maybe you’ve started on your SD-WAN path, are building out support for IoT connected devices, and are eyeing machine learning to aid in understanding the volume and velocity of data traversing your networks.

Ask yourself if you can have a unified view across the network to monitor all those tools, devices, applications, and the end users accessing them. You probably have insight into tools and applications on an individual or even a group level. And when there’s an outage, compromised SLA, or a trouble ticket is logged, you can drill down to find out the source of the issue. But an already taxed network management team doesn’t have time for that, nor does the customer or employee on the other end of the network issue.

What often happens is that the fire is extinguished but the root cause isn’t completely resolved. Or, in some instances, the newest addition to the network is viewed as the culprit. For example, a large healthcare company recently had a network performance issue and was quick to point fingers at the SD-WAN because it was the newest addition. A closer look across the network showed the issue had nothing to do with the SD-WAN. Without that unified visibility, knee-jerk reactions could have had bigger repercussions across the company.

Just like the emergence of cloud computing, it’s not always clear how sweeping technology innovations will impact the network. However, when you have complete and unified visibility, you futureproof your technology investments by being prepared for the next big thing. These days, that’s the emergence of software defined networks.

[1] 2018 Gartner Magic Quadrant for Network Performance Monitoring and Diagnostics.