The Network Must Change, Now

Over my 20 years of building IT infrastructure, I've seen IT professionals morph from jacks-of-all-trades to deep specialists. Indeed, I followed that arc myself. I started as a Novell and Microsoft pro who knew a lot about storage, security, networking, data integrity, and messaging. I did "all the things," so to speak. But over a period of five years, I changed into a networking specialist with a deep focus on data transport. If your data went through a device rather than to a device, that "through" box was probably my responsibility.

Deep specialization has taken me far in my technology career. "The network guy" is a moniker I've worn with pride, happy to be the one who could explain what actually happened inside of the cloud hastily scrawled on the whiteboard to represent a data center fabric. And while all that networking knowledge I've worked so hard to accumulate is valuable, I'm finding it's not enough, not anymore. No more is infrastructure about technology silos. Infrastructure is about application delivery, and the network is being dragged into a new (old) world of unified infrastructure.

Networking finds itself at a complex place. The past is tried-and-true, with reference designs to meet just about any organizational requirement a Google search away. But that past is also one of manual, fragile configuration that is resistant to change. The future of IT is not compatible with a network that waits for a human to make a change in accordance with a complex process that takes weeks. And thus it is that the future of networking becomes important. Yes, we grumpy old network engineers know how to build networks in a reliable, predictable way. But that presumes a reliable, predictable demand from business that just isn't so in many cases.

Organizations seek an IT infrastructure that allows them to deploy applications rapidly. As quickly as developers commit a change, that new code demands to be stood up and accessible to the world. New VLANs? New security rules? Compliance concerns? Monitoring and reporting of network traffic? That had better be part of the publication process, as it's not going to wait for some person in a cubicle to snap the Lego bricks together.

Learn more from Ethan about the Future of Networking at a two-day summit presented by Packet Pushers at Interop Las Vegas this spring. Don't miss out! Register now for Interop, May 2-6, and receive $200 off.

Depending on your perspective, networking's future lies in automation. Networking's future lies in hybrid cloud. Networking's future lies in software-defined changes driven by a policy engine. Networking's future lies in abstracting the hardware away, and making consumption a job for code talking to standard APIs. Or does it?

"What is really going to work?" is the question at the heart of networking's future. Not all trends are going to win. We've already seen SDN devolve into meaningless marketing semantics. Centralized controller architectures look suspiciously like scale and state problems we solved decades ago. And yet, there is progress being made towards making the network programmatically accessible, maybe even in a standardized way. The time to pay attention to these ideas is now, as new networking technology is proving real in many cases, and valuable enough to be useful to IT.

You won't necessarily hear about networking innovations from your incumbent vendor. They are happy to sell you the upgraded version (now with chrome tailpipes!) of what you've been buying for years. You have to reach out to the open source community, startup vendors, and academia to see what's really happening at the edge of networking's future. And you should. There's money at stake -- many innovations are driven by cost concerns. But even beyond that, there's a chance to be a leader. Knowing what's now makes you look solid. Knowing what's next makes you look like a visionary.