Dell Technologies and Microsoft expand private cloud partnership to Nvidia HGX B300 systems for enterprise AI. The package combines Dell infrastructure with the Nvidia platform and Azure Local and Foundry Local software to bring inference of sensitive data closer to the public cloud. Microsoft, however, has not specified any delivery deadline for validated configurations.
Costs depend on intended uses
Many companies are still seeking to identify their future uses of AI and the necessary resources. However, the profitability of owned equipment depends on its utilization rate and anticipated demand. Holger Mueller, vice president and principal analyst at Constellation Research, doubts that reasonable scaling is still possible at the current stage of adoption.
Dave McCarthy, group vice president for cloud and data centers at IDC, believes local compute makes sense for large, consistent, and predictable inference volumes. For large, cutting-edge models operating at a sustained rate, IDC sees a potentially much lower total cost of ownership than the cloud, including electricity, cooling, and premises expenses. Holger Mueller also believes that an organization that knows its needs can often spend less on its own equipment.
Companies retain responsibility for equipment
With the additional control that a local installation brings, comes operating responsibilities. Companies must support GPUs, storage, networking, power and cooling, and software driving models and resources. According to Dave McCarthy, 73% of people surveyed by IDC in June blame the complexity of AI infrastructures for delaying their release into production.
The same analyst believes that the Dell-Microsoft combination eases lifecycle monitoring, firmware coordination and technical support. Holger Mueller credits integrated platforms with reducing deployment work and removing integration risks that customers would otherwise assume. However, IT teams remain responsible for managing capacities, optimizing processing and maintaining the infrastructure.
The evening letterFree · 7 p.m.
AI transforming business?
AI strategies, movements and tricks deciphered, every evening in 5 minutes. Free.
Included upon registration: our selection of the best AI guides & comparisons.
Choose your pace
Free · No spam · Unsubscribe in 1 click
Microsoft validates Dell, HPE and Lenovo
At Ignite on November 17, Microsoft announced that Foundry Local was available on Azure Local and confirmed compatibility of this infrastructure with Nvidia HGX B300. Azure Local hosts workloads in customer-controlled environments; Foundry Local allows models to be deployed and run locally. The hardware extension supports large-scale inference and multi-GPU agentic workloads.
Dell, HPE and Lenovo are among the manufacturers whose systems Microsoft has validated. In the expanded partnership with Dell, Nvidia accelerated computing joins the Azure Local environment through a validated hardware provider.
Hybrid demand goes beyond sovereignty alone
As of June 2026, 85% of companies surveyed by IDC on global digital infrastructures say that their AI processing requires local resources such as the public cloud, reports Dave McCarthy. According to him, priorities are evolving without crowding out the public cloud. Sovereignty and latency weighed heavily on 51% of respondents, compared to 42% for costs.
Steven Dickens, CEO and principal analyst at HyperFrame Research, attributes the hybrid focus to sovereignty and token economics concerns. He points out that open weight models increase deployment possibilities as organizations seek to gain greater control over their data and AI spending. In finance, health and administration, obligations relating to the location and processing of data can influence technical choices.
IDC predicts that the sovereign private cloud will be among the fastest growing segments of the private cloud through 2028. Dave McCarthy describes Azure Local as a platform that can support the rise of sovereign inference on GPU, notably with Dell. Holger Mueller, however, recommends not placing sovereignty at the forefront of reasons for on-site transfer: public cloud providers also offer controls over data. In his eyes, the commercial advantage linked to the execution environment must take precedence.

