Modernize Your Network Performance Monitoring


Network monitoring has never been easy, but it has become far more difficult as the complexity of IT infrastructure has soared.

Part of the challenge is the huge amounts of performance and management data now being generated by growing networks. Another factor is how network monitoring data is accessed continues to evolve. Teams are rearchitecting their networks to accommodate the latest in Wi-Fi, software-defined networking, multi-cloud, SaaS, virtualization, 5G and other technologies.

To keep up, network monitoring systems must be able to collect and analyze data from software-defined control systems and their associated virtual instances and physical devices, and from legacy network resources—at the same time. Complicating this challenge further are entirely new types of network “devices”—like virtual routers—that are being provisioned, and in turn, are producing new types of network data and new ways that data flows throughout the environment.

All of the performance data now being generated for devices and their key indicators needs to be monitored continuously, analyzed quickly and accurately, and when necessary, acted upon rapidly. Doing so with modern networks requires equally modern monitoring approaches.

Yet many networking teams in enterprises, CSPs, and MSPs are still trying to do the job with outdated tools and approaches. They struggle to gain accurate insights from their network data and use them effectively. As a result, their operational efficiency around network monitoring is nowhere near where it could and should be.

This need not be the case. There are steps that networking teams can take to increase their operational efficiency and effectiveness. Since “you can’t manage what you can’t see,” it starts with establishing solid network coverage with comprehensive data collection. Once that foundation is established, teams can take the next steps: overlaying analytics, adding context, building reusable content, and creating problem-solving workflows.

This paper outlines these steps, all of which are enabled by combining comprehensive network monitoring with intuitive ways for users to work with data to gain valuable and actionable performance insights. With those insights, NetOps and IT teams can ensure the consistent, reliable delivery of high-quality network services and applications their organizations depend on today. This paper provides a roadmap for getting there.


The basic task at hand for NetOps and IT teams is to ensure their networks, and the services and applications that run on them, are working properly. In our highly complex networks, however, things sometimes just get jumbled up or simply break. When that happens, NetOps and IT team members shift into the pressure-packed part of their jobs—identifying, locating, diagnosing and fixing problems quickly. In other words, it’s on these teams to deal with network issues before they trip up too many users and bog down the business.

Most teams are still managing hardware-centric data centers with physical, rack-mounted servers, routers and switches hard-wired with legacy, MPLSbased WAN links, etc. Thanks to digital transformation initiatives, they’re now also responsible for managing a whole new networking world. That world includes things like virtualized network services, SaaS and cloud-based architectures that draw on network resources not on-premise but out in the ether somewhere, software-defined WANs, next-generation Wi-Fi, and more.

In these new virtualized, cloud-based and software-driven environments, different things happen, and they happen very quickly. Trying to monitor and manage new infrastructure with a traditional (legacy) network monitoring system presents many challenges. Designed mostly for yesterday’s networking requirements, those systems struggle to keep up with today’s faster and more dynamic networks.

That said, transitions from legacy to new network environments rarely happen overnight. Instead, organizations, especially large ones, take a measured approach and make the transition over time. That sets up the situation in which an organization is running two separate network environments simultaneously. In those scenarios, it’s highly advantageous to have a monitoring system that can straddle both environments.

Given the topic of this paper, however, we will keep its focus on monitoring requirements for modern networks. That ‘must-have’ list includes monitoring capabilities that are just as dynamic, flexible and scalable as the new infrastructures they need to watch over.

But what does that mean, exactly?, for openers, it must be able to handle all of the ‘classic’ monitoring tasks with hardware-based devices, SNMP polling, device flows, etc. Beyond that, a modern network monitoring system needs to be able to collect network and infrastructure metrics from any source – NFV, SDN, SD-WAN, next-gen Wi-Fi, backhaul, 4G/5G, and more – regardless of the size or scale. It must be capable of managing the thousands of different device types that might be present in the network. And it must have the flexibility to quickly add support for new device types as they emerge.

In short, their monitoring solution must be able to collect every performance metric available from the network and digital infrastructure, and integrate those metrics with flow, log and user experience data. These capabilities and the comprehensive coverage they deliver, give NetOps and IT teams a foundation of visibility that is an absolutely critical first step to the rest of this process.

Readers who have digital transformation projects that depend on legacy monitoring systems should rethink their strategies. Slow polling of some fraction of the devices on today’s networks is a sure-fire way to make those projects fail, with potentially catastrophic results. Modern monitoring for modern networks is the answer.

The monitoring solution must be able to collect every performance metric available from the network and digital infrastructure, and integrate those metrics with flow, log and user experience data.


Once a team has conquered this first step and is able to collect all the varied types of data available in their environment with speed at scale, they are ready to tackle the next challenge. That is, making it easy for any technical staff member to make sense of the data. This must extend beyond the seasoned network gurus who are intimately familiar with the characteristics, behaviors, and up- and downstream relationships of every device in their environments.

It has to be easy for other, ‘mere mortals’ to be able to quickly see and understand what the data is telling them. That comes with purpose-built analytics. They also need to be able to determine what to do about the insights provided by data and analytics. That comes with customizable visualizations. Last, they need fast and easy ways to confirm the best course of action for issue resolution. That is unlocked by reuseable, problem-solving workflows.

These are the key capabilities that need to be in place for NetOps and IT teams to increase their network monitoring operational efficiency. Each of these areas is described in more detail below.

Every network monitoring solution enables a NOC staff member to look at a particular device such as a core router, and create a report showing, for example, that router’s memory utilization. But that report, in and of itself, is not that useful, especially if that particular staffer doesn’t happen to be familiar with that device.

Let’s say the report shows that there has been a sharp increase in memory usage by that server over the past 24 hours. Is that normal behavior? Is it within an acceptable range? Are things good, bad or about to get ugly? Our NOC person, who is filling in for an expert colleague who’s out sick or on vacation, looking at that single report, has no idea whether what he or she is seeing is okay or about to become catastrophic.

To read full download the whitepaper:
Modernize Your Network Performance Monitoring


Previous articleTop Five Monitoring Use Cases For G Suite
Next articleThe Essential Guide to Data Preparation