Nexus HyperFabric: connecting the parts of an AI data centre
Archive focus: June 2024 · Prepared 17 October 2026 · XConnect FZE
Retrospective: This article was prepared on 17 October 2026 for our June 2024 archive. It reviews the announcement and its infrastructure planning implications; product plans described at the time may have changed.
The development
At Cisco Live on 4 June 2024, Cisco announced Nexus HyperFabric AI clusters with NVIDIA. The proposed solution combined Cisco networking, NVIDIA accelerated computing and AI software, and a VAST data platform. Cisco presented it as a way to simplify how customers build and manage infrastructure for generative AI.
Practical implications
The following is XConnect’s planning interpretation.
An integrated cluster can make procurement easier to understand, but the application still determines the design. A useful evaluation starts with the data pipeline, the model workload and the people who will operate the system. Capacity planning must also consider where equipment will be installed and how it will be maintained.
What to review
- Confirm rack, power and cooling requirements alongside computing specifications.
- Identify the owner of each layer, including storage and model-serving software.
- Agree acceptance tests for throughput, recovery and day-to-day administration.
Explore our network solutions and data centre solutions, or contact XConnect to discuss your site requirements.
Source: Original announcement — investor.cisco.com. Reviewed 17 October 2026.