Data center networking is a discipline built on hardware and embedded software but differentiated through operations. The most advanced networking shops in the world certainly feature some of the highest-scale solutions, but that scale is only possible because of their operational investments.

It's no surprise, then, that most of our industry's really transformational innovation centers around operations and features new and better ways of managing networks, rather than merely building them. Whether it's automation and DevOps, hybrid and multicloud, or intent-based networking and AIOps, there are many emergent capabilities poised to elevate operations and bring networking more in line with advancements that have already taken root on the compute and storage side.

But with many versions of management systems now featuring these capabilities, how can you tell the difference between the flashy and truly powerful platforms?

The big picture

Years ago, I spoke with an online retailer that had a cloud practice and a particularly advanced operations shop. I asked how long it took them to stand up a new application, expecting to hear that the process was highly automated and that they could do it in minutes. The actual answer? Six to eight weeks. I was surprised, but when I heard the reason why, it made sense.

The actual time to make the changes was measured in seconds. They had automated that part. But their proposed changes had to be staged in a workflow system. That triggered a set of reviews and activities, which then led to permission to execute within a change window that was scheduled monthly.

The key point? The elapsed time—the time it took them to make the changes—wasn’t dominated by the time it took to execute the commands. It wasn’t even dominated by the networking-only parts of the workflow.

If we want to remove time from the system and speed everything up, we have to expand our view beyond the network.

This is a hugely impactful conclusion to reach. It means that while our automation ambitions might start with familiar networking tasks, we need to consider the broader landscape in which those tasks exist.

Automation is a function of the broader ecosystem

When operations teams ask me about how to identify these workflows in a broader context, the advice I give them is straightforward. Pick any tool your team uses. With very few exceptions, each of these tools represents a window to a workflow. The tool will see or detect something, or maybe allow for some action to be taken. Whatever the role, that activity is part of some broader workflow.

So if we want to tackle more meaningful workflows—and those where elapsed time tends to accumulate—we need to be on the lookout for the intersection of these tools, their associated workflows and the networking workflows that are already familiar to us.

The important implication here is that the automation surface area is not defined by a single tool or workflow. Rather, it's a function of the broader ecosystem of tools. This means that the management platforms that will inherently drive the most value will be those that consider integrations with surrounding systems.

Ease of integration matters most

If integrations are the key, then the things you look for to understand whether a product is flashy or meaningful should change. The UI matters, but the way tools are integrated is the truly telling characteristic. What APIs exist? How is data normalized? Are interfaces versioned and maintained across different releases? Can you create complex dashboards that pull things together from different sources using no-code models that don't require source access to contextualize to your environment? How are workflows strung together into more complex operations?

By changing your focus, you can start to evaluate these platforms based on how well they integrate rather than on how snazzy the time series database interface is.

Of course, things like look and feel matter, but anyone who wants to scale their operations will realize that the UI might not even be the dominant consumption model over time. Is your team looking to click their way through to completion? Or is it more likely that you are triggering a workflow from ServiceNow without requiring some human intervention?

Learnings from the cloud

The compute and storage worlds have already figured this out. Kubernetes is the de facto operating system for the cloud. The way it handles resource management and workflow automation is instructive.

The same principles ought to translate to networking, not just because the most meaningful workflows will span compute and networking, but because networking's next-generation workforce is more likely to be well-versed in Kubernetes and cloud than vendor-specific CLI and antiquated network operations models.

Wherever you are in this discovery process, let me offer some simple advice: Expand your purview from the network to the ecosystem and evaluate your options in the context of that ecosystem. When you do that effectively, you should know which solutions are attractive but incremental and which are likely to create more durable value for you and your organization.

Adaptable data center network operations

Building on these principles, Nokia has launched a new data center network automation platform called Event-Driven Automation (EDA), which is compatible with open-source cloud-native technologies and easily integrates with a rich ecosystem of tools and clouds.

This means EDA can be used with a variety of surrounding tools—including a wide range of IT service management systems, event notification systems, cloud management platforms and Kubernetes toolchains—without being restricted to proprietary platforms. This enables you to speed up the creation and deployment of solutions that will meet evolving end-user demands.