How to succeed with data center network migration
Our industry is in the throes of one of the most dynamic investment cycles in all of recorded history, and there is an awful lot of noise. That noise can make it nearly impossible to discern fact from fiction and theory from practice.
Understanding the practice lets us get things done. For a recent Total Network Operations podcast and SReXperts event plenary, I discussed the considerations of practice in data center networking with Scott Robohn, CEO and Co-Founder, Solutional, and Pushpendra Pandey, Head, Global Network Services, Nokia.
Pushpendra is currently leading Nokia through a network migration effort in 17 global data centers that support 75,000-plus employees working remotely and in more than 200 offices. He started with one data center site that supports a large manufacturing operation. After the migration, he saw an 80% reduction in incidents at that site. Now, the project continues to raise reliability everywhere.
Scott, Pushpendra and I discuss the project in depth in the podcast. Pushpendra and his team really understand why they are migrating and where they want to go.
Why data center network reliability matters
At a large corporate entity—especially one that serves the largest service providers in the world, which collectively have billions of customers—expectations for reliability are high. The IT team must deliver service levels of five nines or better and support all the adds, moves and changes common to any large corporation.
Before Pushpendra and his team began the data center network migration project, efforts to achieve these levels were taking too much time. They were also falling short too often, and it was impacting key parts of our business, including our 24x7 manufacturing operations.
Unfortunately, many IT teams are all too familiar with failing and poor-performing applications. A summer 2025 survey by Futurum found that reliability is the #1 decision criteria for senior IT and infrastructure leaders when it comes to choosing data center solutions. You can read about these and other findings in Futurum’s market report, The Data Center Networking Imperative: Key Trends Driving the Next Era of Data Centers.
A quest for simplicity and time
More than changing the current situation, Pushpendra and his team had a vision of where they wanted to go. They looked for modern solutions that would make network operations more predictable and give time back to the team. They wanted a solution that would be simple to deploy, configure and operate, so that they could easily do small migrations, change infrastructure and upgrade software. They were looking for solutions that would keep experts from having to do repetitive work.
Pushpendra told us the key to ongoing success for his team is to stop doing the repetitive tasks machines can do and start doing more intelligent things. He acknowledged that the next generation of workers is now growing up with AI. If you tell them to start creating VLANs for an interface or monitoring alarms because a port went down, they are just not going to work for your company.
Reducing problems, burnout and risk
He also talked about the reality of burnout in operations teams. A Nokia employee survey revealed that workers were frustrated by cases where trouble occurred late in an eight-hour shift and compelled them to work overtime until it was resolved. Most often, it turned out that the root cause was an easy and common problem. The workers wanted better tools that could prevent it from happening at all.
These tools are available today. Pushpendra’s team is adopting the Nokia Event-Driven Automation (EDA) platform for its abstraction capabilities, programmability and digital twin functionality. For the migration project, the team is thinking about how to improve network and application reliability, as well as how to make things better for employees and customers.
Pushpendra also spoke about the risk associated with change during the podcast and at our event. When the team assessed the risks, it found that there was more risk in sticking with the status quo where outages were unpredictable compared to embracing the calculated and managed risks of migrating the data centers and their legacy of workloads. Running the old data center network had just become too complex. It was time for a change.
A new way to measure data center migration success
If you’re planning a data center migration, I think it's worth following the Nokia IT team’s approach and having deep conversations about vision and strategy. Indeed, if you have no vision, any step is the right direction. But events like SReXperts that ground that vision in practical activities help people get started. And ultimately, success will be measured in lived experiences more than theorized outcomes.
And if you're interested to learn more, you can listen to my conversation with Scott and Pushpendra on TNO044: Inside a Global Enterprise Data Center Network Migration at the Total Network Operations podcast.
