Nvidia's next multibillion-dollar market lies outside the data center, analysis says
Nvidia posted $96.2B quarterly revenue, but analysts see its next growth frontier in distributed AI compute at the edge, using high-speed fabrics to overcome power density limits.
Chief executive Jensen Huang said in a statement: “AI has reached its inflection point. It’s doing useful work. Its tokens are productive and profitable. Now, compute is revenue. The AI infrastructure buildout is at full steam. Vera Rubin, now in full production, was built to power exactly this moment.”
The analysis points to a physical constraint: modern NVLink-scale architectures demand up to 140 kW of power density inside a single rack. While that is manageable for purpose-built mega-clusters, the existing edge — telecom central offices, industrial campuses and enterprise data centers — is typically capped at 30 kW to 50 kW per rack, making it unable to handle the weight, cooling or power of a modern AI rack.
To bridge the gap, infrastructure architects are disaggregating the compute envelope across four or five 30 kW racks and linking them with high-speed optical scale-up fabrics, so adjacent physical nodes behave as one low-latency AI system. This lets operators deploy AI without tearing down or retrofitting buildings.
According to the analysis, this is a classic disruption strategy: rather than displacing hyperscale infrastructure, the new architecture creates net-new enterprise and distributed deployments. It estimates the installed edge footprint holds about 30 gigawatts of aggregate power capacity, which is fragmented but available.
The approach unlocks new markets, including a telco pivot from transport to intelligent services by injecting disaggregated AI compute into regional central offices.