MUST READ: Machine Learning for Network and Cloud Engineers

Javier Antich, the author of the fantastic AI/ML in Networking webinar, spent years writing the Machine Learning for Network and Cloud Engineers book that is now available in paperback and Kindle format.

I’ve seen a final draft of the book and it’s definitely worth reading. You should also invest some time into testing the scenarios Javier created. Here’s what I wrote in the foreword:

Artificial Intelligence (AI) has been around for decades. It was one of the exciting emerging (and overhyped) topics when I attended university in the late 1980s. Like today, the hype failed to deliver, resulting in long, long AI winter.

The situation has changed drastically in recent years. AI solutions can recognize images, play chess or Go, or try to appear semi-intelligent in casual conversation. You can guess what the results of this unexpected progress might be: if you want to get your startup funded (or your project approved), it must mention artificial intelligence or machine learning regardless of whether that makes sense or not, resulting in another wave of mostly misguided hype. The networking industry is no exception – every vendor is talking about artificial intelligence, even if their product contains nothing more than a long list of if-then-else decisions.

Does that mean we can safely ignore the AI/ML hype and keep doing what we have been doing for the last 50 years? Of course not. AI/ML technologies are at a point where they are worth considering outside of academic environment, but not without a hefty dose of critical thinking. The next obvious question is thus, “where could I get the unbiased information I need to evaluate whether AI/ML makes sense for the problem I’m trying to solve?”

A few years back, I got an out-of-the-blue email from Javier Antich arguing about intricate details of identifying malicious domains from DNS traffic. He was clearly an expert with a deep understanding of AI technologies (beyond the currently-hip neural networks) and their networking applications. He was kind enough to create an webinar with a very appropriate title — AI/ML: The Good, the Bad, and the Ugly – in which he discussed AI-related technologies, their use cases, advantages, and drawbacks. Not surprisingly, that webinar has been a huge success receiving tons of accolades.

Most webinars have just one tiny problem: discussing all the relevant quirks and nuances is usually a Mission Impossible due to time limitations. Fortunately, Javier solved that problem and wrote a deep-dive book (that you obviously own – congratulations) covering the same set of topics in way more detail. While it’s always good to have a detailed discussion of multiple related technologies, he went a step further and created almost a dozen scenarios, from detecting outliers and anomalies in network behavior to clustering (what devices do I need), forecasting (how many subscribers will I have) and even “natural” language processing to analyze router configurations. To make the scenarios even more realistic, he created a network simulation tool, and published all the source code on GitHub, allowing you to try them out with little extra effort beyond installing the necessary software.

I’m pretty positive that you’ll find at least one of his use cases applicable to your network, and when you do, dig deep into the technology chapters to figure out why it works and how to adapt it to your environment. I wish you good luck in exploring the AI/ML world and hope you’ll find Javier’s book a worthy companion.

Add comment