What Is Network Agility and Why Does It Matter?

In a 2019 Top Trends Transforming Network Operations survey, 34% of networking pros identified improving network agility as their top business goal for the year.

The stat isn’t surprising. “Network agility” is considered the future of networking, but the term itself has become a bit of a buzzword. Everyone’s talking about it, but no one can agree on its definition.

So what does network agility actually mean? We surfed the web and analyzed what (almost) everyone has to say. And now we’ve broken it down into an easy to understand, (almost) jargon-free explanation—and answered some questions you may have on the topic.

What’s network agility?

Network agility is the network’s ability to respond to network changes in real-time while keeping pace with a business’s evolving technology needs.

An agile network can adapt to changes—like an uptick in traffic or an influx of new devices—as they happen while remaining flexible, secure, and easy to manage.

For a network to be considered agile, it needs to have three distinct characteristics:

• It can scale quickly: In a normal network, the rules and configurations needed to expand a network are coded by hand. In an agile network, scaling up is a hands-free process where a computer can create and deploy network templates to add new devices to the network and reconfigure existing devices. For this to happen seamlessly, the network needs fewer vendors and a logical layout.
• It has complete visibility into the network: In an agile network, data is everything. All data from endpoints, network devices, and traffic flows—including performance data, alerts, and more—needs to be collected and stored. The data is then analyzed using machine learning and artificial intelligence to be used in maintenance and troubleshooting.
• It’s completely autonomous: Using the data collected, an agile network will learn root causes behind specific alerts and the appropriate steps to troubleshoot them. It will try each of those steps until it’s successful and further refine its ability to identify and remediate problems. Eventually, the network will be able to find and fix issues on its own to keep things running smoothly—no human input necessary.

If you’re thinking this sounds a lot like intent-based networking (IBN)—where you can identify your networking intent and software figures out the best way to achieve it—you’re right. The two concepts share several similarities.

The biggest difference between the two is IBN is attainable now. Network agility, on the other hand, is something the networking world is still working towards. While elements of it are attainable, it’s still just out of reach.

How can I prepare my client networks for network agility?

Unless your client networks were built using modern architectures and technologies (and with the concept of network agility in mind), then they’re likely mishmashes of different devices (made by different vendors).

If that sounds familiar, then achieving network agility will require major modifications to your client networks and to your work processes. But it doesn’t have to be done in one go. Instead, there are four proactive steps you can take to prepare your networks:

1. Sell clients on standardization: A network with fewer vendors is easier—for you and a super computer—to monitor and manage than a diverse network. It’s also easier to scale as the network evolves over time. The challenge is standardization is costly and time-consuming, but with a strong “take it or leave it” pitch focusing on your team’s expertise with a specific vendor, it can be done.
2. Be strategic in network planning: When updating or adding to the network, you need to consider if the change is a logical extension or a hack. If you can’t answer the question “Does this extension set the network up for success in the future?” with a resounding “yes,” you’re building a network that will be difficult to fully see and automate.
3. Document everything: Topology maps, device inventory, alerts, troubleshooting efforts—it doesn’t matter what it is, you should document it (if your rebuttal here is “I don’t have time,” don’t worry: You can automate this process by integrating your PSA, RMM, and network management tools). This information is important for the next step.
4. Create tight-knit processes: Using your documentation, you can create helpful processes for your team when problems arise. When a specific alert triggers a ticket, your techs can refer to the document and see which steps were taken to fix it. Not only will this effort boost your efficiency now, it will also help with machine learning one day.

Why should I start preparing for network agility now?

While we’re still a ways away from achieving true network agility, it’ll be here before we know it—and if your client networks aren’t ready for when it arrives, you might find yourself in some hot water.

The fact is many MSPs are already taking steps the very steps we’ve outlined to prepare their client networks for the future. If you don’t start now, then you’ll be scrambling to modify all of your client networks at once, which is a big task that requires a lot of money and manpower.

And while you’re focusing all of your resources on that, other MSPs will be able to sell your prospects and clients on their next-level troubleshooting efficiency—and their ability to keep up with networking trends.

That’s not good for business. So what are you waiting for?