Network equipment manufacturers like Cisco are thriving due to the financial benefits of AI-powered network systems
The market for AI networking is experiencing a rapid growth, mirroring the production rate of GPUs by tech giants like Nvidia and AMD. According to Dell'Oro Group, AI networks are expected to drive nearly $80 billion in network sales by 2030, potentially avoiding an AI bubble burst.
In the realm of AI data center market, revenue is projected to increase from approximately $236 billion in 2025 to around $934 billion by 2030, at a CAGR of 31.6%. This growth reflects the high demand for high-performance networking infrastructure to support AI workloads.
The AI platforms market, covering software and services that enable AI deployment and networking, is forecasted to grow from about $11.3 billion in 2024 to $56.3 billion by 2030, with a CAGR of roughly 30.8%. This growth is expected to continue, with projections extending to nearly $94.3 billion by 2030 at a CAGR of approximately 38.9%.
The broader AI infrastructure market, which includes networking hardware along with compute and storage, is expected to rise from around $60 billion in 2025 to $156 billion by 2030. This market is projected to nearly reach $500 billion by 2034, growing at about 26.6% CAGR.
Key sectors driving this growth include healthcare, finance, manufacturing, and technology companies investing significantly in AI-optimized data centers and platforms.
In terms of specific companies, leading cloud providers like Microsoft Azure, Amazon Web Services (AWS), and Google Cloud are offering scalable AI solutions and hyperscale data center infrastructure. Hardware chip producers specializing in AI accelerators such as GPUs and TPUs are also expected to benefit, with industry leaders like Nvidia and Google’s TPU division implied.
Companies focusing on green AI data centers and energy-efficient networking infrastructure will also gain opportunities due to the rising emphasis on sustainability.
Notable developments include Nvidia's acquisition of Mellanox for $6.9 billion in 2019, which has positioned the company as one of the biggest names in AI networking. Nvidia is also developing photonic switches to cut down on the number of optical transceivers required in AI networks.
In Q1 of 2025, Nvidia reported network sales up 64% quarter over quarter to roughly $5 billion, including sales of InfiniBand switches and ConnectX NICs. These orders are more than double the company's original $1 billion target stated in Q4 of fiscal year 2024.
For a 128,000 GPU cluster, the switch count could be as high as 10,000 if older 25.5Tb/s network kit is used. However, with modern 51.2Tb/s switches and 400GbE links, only a leaf and spine layer are required, reducing the switch count significantly. For clusters exceeding 8,192 GPUs, a three-tier fat tree topology is required to provide full-non-blocking connectivity for the AI back-end networks.
Hyperscalers aren't particularly picky about who or where the AI network kit comes from, as long as it works. This opens up opportunities for various networking vendors to capitalize on this growing market.
Arista's CEO, Jayshree Ulla, expects aggregate AI networking revenue to be ahead of $1.5 billion in 2025 and growing in the following years. Arista forecasts full year revenues of $8.75 billion, with AI-related sales expected to account for approximately 17%.
Juniper Networks is presumably experiencing high demand for AI infrastructure, though the impact is harder to assess due to its merger with HPE. Cisco's AI infrastructure orders exceeded $800 million during the fourth quarter of 2025, and $2 billion for the entire fiscal year. Cisco is optimistic about its opportunities in traditional enterprises for AI infrastructure, where it can upsell customers on value ads, service contracts, and software subscriptions. Optics accounted for roughly a third of Cisco's $800 million AI networking revenues in Q4.
The convergence of back-end and front-end in AI networking is expected, making it difficult to parse the two in the future. However, one thing is clear: the AI networking market is set for robust multi-billion-dollar growth driven by the increasing adoption of AI across industries and the scaling of AI-optimized data centers by leading technology firms.
- The AI data center market, driven by high-performance networking infrastructure demand for AI workloads, is projected to increase from $236 billion in 2025 to $934 billion by 2030.
- The broader AI infrastructure market, including networking hardware, compute, and storage, is expected to rise from around $60 billion in 2025 to $156 billion by 2030 and could potentially reach $500 billion by 2034.
- Leading cloud providers like Microsoft Azure, Amazon Web Services (AWS), and Google Cloud are offering scalable AI solutions and hyperscale data center infrastructure.
- Nvidia's acquisition of Mellanox in 2019 has positioned the company as a significant player in the AI networking market, with network sales up 64% quarter over quarter to approximately $5 billion in Q1 of 2025.
