Nvidia's Hidden Giant: Networking Division Soars Past $31B

The tech titan's data center networking business is quietly becoming a revenue powerhouse, challenging its famous chip division.

Nvidia's data center networking division, fueled by a 2020 acquisition, has rapidly grown into the company's second-largest revenue stream. Reporting over $31 billion annually, this segment is now a critical component for building 'AI factories,' connecting GPUs and powering advanced AI models.

Mark Ellison

By Mark Ellison

March 19, 2026

4 min read

Nvidia's Hidden Giant: Networking Division Soars Past $31B

Key Facts

  • Nvidia's networking business reported $11 billion in revenue last quarter, a 267% year-over-year increase.
  • The division brought in over $31 billion for the full year.
  • It is now Nvidia's second-largest revenue driver, behind its compute business.
  • The networking business includes technologies like NVLink and InfiniBand Switches.
  • Its origin traces back to the 2020 acquisition of Mellanox for $7 billion.

Why You Care

What if a tech giant had a secret weapon, one almost as as its most famous product? Nvidia, known globally for its AI chips, is doing just that. They are quietly building a massive business that could soon rival its core chip operations. This isn’t just about corporate growth; it’s about the very infrastructure powering the AI revolution. How will this impact your digital world?

What Actually Happened

Nvidia’s data center networking division has seen extraordinary growth, according to the announcement. This segment, designed to connect data centers, has quickly become the company’s second-largest revenue driver. It now sits right behind its dominant compute business. Last quarter, this division reported a staggering $11 billion in revenue, representing a 267% year-over-year increase. For the full year, it brought in more than $31 billion, the company reports.

This rapid expansion is largely driven by the increasing demand for AI processing. The networking business includes crucial technologies like NVLink, which facilitates communication between GPUs (Graphics Processing Units) within a data center rack. It also features Nvidia InfiniBand Switches, an in-network computing system. These components are essential for creating what Nvidia calls an “AI factory”—a data center specifically for training AI models.

Why This Matters to You

This rapid growth in Nvidia’s networking business has direct implications for the future of AI and, by extension, your daily life. Think of it as the superhighway for AI brains. Without efficient networking, even the most AI chips can’t communicate effectively. This means slower AI creation and less applications for you.

For example, imagine you’re using a generative AI tool to create complex images or write detailed reports. The speed and accuracy of that tool depend heavily on the underlying data center infrastructure. Nvidia’s networking solutions ensure these AI factories run at peak performance. This translates to faster, more capable AI services for your personal and professional use.

Kevin Cook, a senior equity strategist at Zacks Investment Research, highlighted the scale of this achievement. He stated, “[Nvidia’s networking business] reports $11 billion for the quarter; that number is greater than Cisco’s networking business, almost as big as the full-year estimates.” This shows just how quickly Nvidia has scaled this crucial segment. Are you ready for even more AI tools in your hands?

Here’s a quick look at the division’s performance:

MetricLast QuarterFull Year
Revenue$11 Billion$31 Billion+
Year-over-year Growth267%N/A
Ranking2nd Largest SegmentN/A

The Surprising Finding

Here’s the twist: despite its colossal financial success, Nvidia’s networking business doesn’t receive the same public recognition as its AI chips. The company’s chip business is significantly larger, and even its gaming division, though smaller, garners more fanfare. This segment, which brought in over $31 billion for the full year, operates largely under the radar. It’s surprising because its quarterly revenue alone surpasses the entire networking business of a long-established player like Cisco, as the team revealed. This challenges the common assumption that only the ‘flashy’ components of AI get attention. It shows the immense value in the less visible, yet essential, infrastructure.

What Happens Next

Nvidia’s strategic focus on data center networking, stemming from its 2020 acquisition of Mellanox for $7 billion, is set to continue its ascent. We can expect to see further integration and creation of technologies like NVLink and InfiniBand Switches throughout 2026 and beyond. This will enable the construction of even more “AI factories.” For example, future applications could involve vast networks of GPUs collaboratively training a single, massive AI model in real-time, significantly accelerating research and creation.

Industry implications are substantial. Other companies building AI infrastructure will likely need to adopt or compete with similar networking solutions. For you, this means a future where AI-powered services become even more responsive and capable. Our advice: keep an eye on these infrastructure developments. They are the backbone of the AI revolution. As the company reports, this quiet giant is not slowing down.

Ready to start creating?

Create Voiceover

Transcribe Speech

Create Dialogues

Create Visuals

Clone a Voice