One of my favorite quotes – from Clifford Stoll – is, “Data is not information. Information is not knowledge. Knowledge is not understanding. And, understanding is not wisdom.” This captures the essence of what’s wrong with network management today.

IT is inundated with raw data from logs and alerts, but this mass of data obscures IT’s understanding of what’s wrong. The wisdom that IT departments seek – understanding the health of their network and determining the next steps for optimal operation – is often obscured by an overwhelming volume of data. To cope, IT assigns highly skilled operators to sift through this torrent of data – a completely unsustainable approach.

This, of course, is precisely the kind of problem AI is built to help with. 2024 is shaping up to be the year AI comes into its own with network management, giving networks a ‘voice’ and enabling more efficient and insightful monitoring. Here are the five ways AI will turn raw data into network management wisdom in 2024:

  1. Real-time Insights: Bad actors count on the “fog of war” – the lack of visibility massive amounts of raw data creates. This fog-of-war is why it often takes days or weeks to discern what happened after a breach.

AI can process vast volumes of data in real-time to provide real-time – and actionable – insights. Imagine the value of knowing a ransomware attack is just starting, versus finding out after all your files have been encrypted.

  1. Anomaly Detection: AI can discern what constitutes normal network behavior and identify anomalies. For example, suppose a trusted user is copying a file for which the user has full rights. This would raise no alarm bells. 

But what if the user does so at an hour he or she has never worked? And from a new location? And what if, after copying the file to a central location, a different user – again, a trusted insider – copies the file to an off-network location.

AI can correlate these actions and see a composite threat comprised of a series of seemingly innocuous events. Yes, a seasoned IT pro might make the same conclusion, but perhaps not at 3:00 am on a Wednesday morning.

  1. Automated Analysis: AI can process data from network monitoring tools to pinpoint patterns, trends and anomalies that suggest a problem. This significantly reduces the time required to identify and respond to network issues.
  1. Enhanced Security: AI bolsters network security by identifying and responding more quickly and accurately to threats. Machine learning algorithms can draw on past security incidents to better recognize and react to signs of malicious activity. They can also automate responses to certain threats, allowing IT staff to focus on more complex security issues.
  1. Predictive Maintenance: AI’s ability to predict potential network failures or performance degradation before they occur is another significant benefit. By analyzing historical data and understanding patterns, AI can forecast when a component will likely fail or when network traffic will likely spike, allowing for proactive maintenance or capacity planning.

How CxOs Can Prepare for AI-Driven Network Management

The key to enabling AI is to make the right data available to the full AI ecosystem. Let’s explore this requirement. First, where does the data live that will fuel the network management AI models. The edge?  Data center? In the cloud? The answer, of course, is all three. The log files, traces, etc., live where network traffic does – everywhere.

On the other hand, the AI ecosystem lives in the cloud. This is where the GPU-driven compute lives, AI modeling services live, and smart AI consultants work.

And, once these AI models are built, where will IT deploy them? Once again, we’re back to “everywhere.”

What this means is your network must connect everything you own to everything you own. That includes edge, core and cloud. But it also includes partner networks, customer networks and thousands of remote workers.

To do this, you’ll need a networking platform that combines simplicity, agility, performance, security and reliability. If that doesn’t sound like MPLS or SD-WAN to you, you’re in good company. In a 2023 survey, IT professionals and network architects gave both networking technologies low marks, and indicated a preference for Networking-as-a-Service.

So, by 2024, AI will transform network management, turning vast amounts of data into valuable insights and enabling real-time responses, anomaly detection, automated analysis, improved security and predictive maintenance. To prepare, you’ll want to ensure your network is up to the task.