Enterprise AI infrastructure starts with a complete cabling plan
Archive focus: February 2024 · Prepared 17 October 2026 · XConnect FZE
Retrospective: This article was prepared on 17 October 2026 for our February 2024 archive. It reviews the announcement and its infrastructure planning implications; product plans described at the time may have changed.
The development
On 6 February 2024, Cisco and NVIDIA announced plans to simplify the deployment and management of enterprise AI infrastructure. The collaboration covered Ethernet networking, computing, software and operational support, with options spanning on-premises and cloud environments. The announcement also described NVIDIA Tensor Core GPUs in Cisco’s M7 UCS server generation.
Practical implications
The following is XConnect’s planning interpretation.
For a business evaluating AI, computing capacity must be matched with a workable physical network. The cabling schedule needs to follow the selected topology, rack layout and equipment interfaces. Treat the first installation as a coordinated infrastructure project: the network and facilities teams should review the same drawings before procurement, with a clear owner for changes and acceptance.
What to review
- Define server, storage and switch connections before ordering cabling.
- Ask suppliers for a validated equipment, optics and cabling schedule.
- Pilot a representative rack and document acceptance before repeating the design.
Explore our network solutions and data centre solutions, or contact XConnect to discuss your site requirements.
Source: Original announcement — newsroom.cisco.com. Reviewed 17 October 2026.