I’m excited to share that I’m in the process of writing a book for O’Reilly Media! You can check out the Introduction and first chapter of “A Practical Guide to Modern Networking Telemetry” now. That link leads to an early release version, which means it’s been “lightly edited”. If you have comments, corrections, questions, or even kudos, let our editor know: [email protected].
I’m co-authoring it with Avi Freedman, CEO of Kentik, and it’s a labor of love for both of us. As we wrote in the “About the Authors” section, the idea to write this arose first and foremost to get down on paper all of those samples, examples, stories, and lessons we usually share in at conferences, or in talks, or when helping customers; but also to provide a resource to the readers themselves: who might need to articulate those same lessons to colleagues, managers, or the odd (very odd) person at a dinner party.
There’s one comment from the introduction that I wanted to unwind a little bit in this blog. In the section titled, “What IS the Network, Exactly”, I wrote:
Back in 2023, in his book “The Ultimate Guide to Network Observability“, Avi wrote,
“People often refer to “the network” in their organizations, but in most cases the network isn’t one entity. It’s a complex, diverse, fragmented, and loosely interconnected set of physical and virtual links and equipment, and it’s housed in a variety of places, including data centers, corporate wide area networks (WANs), private and public clouds, the internet, container environments, and even inside hosts. Organizations own and control some of those resources but simply pay to use others.”
Since that time, things remain largely the same. Sure, the cloud has gotten more cloudy (meaning opaque to users) even as it has become more ubiquitous within organizations large and small. And at the same time corporate LAN and WAN environments have not, by and large, become less complex (or sprawling, or expensive). In fact, the opposite has been true. Even as company’s investment in cloud has increased, so too has their investment in on-premises networking gone up.
Why is that? Because the more we – businesses and IT professionals alike – use the cloud, the better we understand what the cloud should be used for; and conversely, where cloud should absolutely not be used, whether due to considerations of cost, complexity, or security. And in those cases, we have brought systems, applications, data, and solutions back into our data centers, and found new and creative ways to connect and combine it with the cloud infrastructure.
And that – the cursed franken-network of modern IT: infrastructure that spans both cloud and on-prem environments, offering all the complexity, cost, and issues of both, while offering very little of the benefits.
The sheer horrific nature of hybrid networking might not be intuitively obvious to folks reading this because we live and breathe in a technical context every day. So let me draw a parallel that might make it clearer: hybrid cars.
Blending a traditional gas-powered combustion engine with a modern fully-electric motor sounds like a great idea on paper, but in practice it’s a nightmare. The design burdens the user with all the quirks and complexities of both systems; it also creates a completely new set of quirks and complexities on top of those; and offers only modest benefits – which are usually completely outweighed by the negative aspects of smushing the two technologies into one form factor.
And that’s what hybrid networks are: the combined costs and complexities of on-premises and cloud networks, with a whole new layer of the same incurred by trying to make them play nicely together.
The big difference between hybrid cloud and hybrid cars is, of course, that we IT practitioners don’t have much of a choice. As much as we wish – hell, as much as we TRIED over the last decade to push environments to fully-cloud or fully-on prem, hybrid is the reality we’re going to live in for the foreseeable future.
And that means we – IT practitioners in general and folks who care about (and are responsible for) monitoring and observability in particular – have to improve the ways we ensure performance and reliability of those environments.
And here, approximately 700 words into this blog, I’ve finally gotten to the problem I hope “The Practical Guide” will help fix:
- the folks who are often responsible for building the cloud-based side of those environments often know very little about networking
- and meanwhile, the teams who are often responsible for building the on-prem side of those environments often know very little about modern observability.
The book offers a clear and complete explanation of everything from the types of devices that comprise a “network”; to the types of telemetry those devices emit; and continues from there to explain how to collect, transform, transmit, and use that telemetry.As I mentioned at the start of this blog, chapter 1 is available in its early release version now. Go ahead and give it a look, and let me know what you think – either in the comments below, or by emailing my editor Gary at [email protected].