Blog Post

Azure Arc Blog
5 MIN READ

What's new with Sovereign Private Cloud at Microsoft Ignite 2026

Bcaldas's avatar
Bcaldas
Icon for Microsoft rankMicrosoft
Nov 17, 2026

As organizations are scaling AI adoption at an unprecedented pace, they need to stay in control of their data, operations, and access as regulatory and operational requirements evolve. Microsoft Sovereign Cloud gives them that control, with choice across public and private cloud environments. At Microsoft Frontier Tour last month, we highlighted how Sovereign Private Cloud, built on Azure Local, brings Azure infrastructure, services, and AI capabilities in customer-controlled environments. 

At Microsoft Ignite 2026, we're building on that foundation with the general availability of Foundry Local, confidential virtual machines, and GitHub Enterprise Local, on Azure Local, along with public previews for agentic AI, connected to disconnected operations, and local data services. These announcements expand what organizations can build and run in their own sovereign environments, while keeping control over where data resides, who can access it, and how systems are operated, even when fully disconnected. 

Sovereign AI in customer-controlled infrastructure

Foundry Local is now generally available on Azure Local. Organizations can deploy and run AI models close to their data, enabling local inference in connected and disconnected environments while maintaining control over their infrastructure, data, and operations.

We’re also expanding Foundry Local beyond model inference to support agentic AI scenarios on Azure Local. This release includes:

  • Model catalog: Run AI models locally from an expanding model catalog, keeping inference close to data, applications and operations.
  • Agentic retrieval: Ground AI in local enterprise data with local ingestion, indexing, and search, plus agents, tool calling, a Model Context Protocol (MCP) server, and a built-on chat experience.
  • Agent service (preview): Build and orchestrate agents, even in fully disconnected environments.
  • Guardrails (preview): Apply guardrails using the guardrail model of their choice.


Support for NVIDIA HGX B300 systems.
As AI workload grow, many organizations need more compute capacity close to their data. Azure Local is expanding hardware support to include NVIDIA HGX B300 systems, so customers can run large-scale, multi-GPU inference, and agentic workloads, including bring-your-own models, on infrastructure they control.

Validated systems from Dell, HPE, and Lenovo will give customers a choice of hardware for scaling AI in their own environments. We’ll share more details as these systems become available.

Confidential computing on Azure Local

Confidential VMs powered by AMD SEV-SNP are now generally available on supported AMD EPYC processors, providing hardware-backed memory encryption and attestation to help protect sensitive workloads while running. We're also expanding hardware choice with Intel TDX-based confidential VMs in preview on Intel Xeon processors.

Resilience and operational independence on Azure Local

We’re enabling organizations to move their Azure Local management control plane without redeploying infrastructure, migrating data, or interrupting running workloads. Virtual machines, AKS clusters, applications, and storage remain in place while management transitions to a new control plane.

The ability to move from connected-to-disconnected deployments (in preview), allows customers to transition existing Azure-connected deployments to locally managed, disconnected environments.

Data services on Azure Local

Building on SQL Server support, we're expanding local data services for modern applications and AI workloads:

  • SQL Server is generally available on Azure Local: Run enterprise database workloads on customer-controlled infrastructure across connected and disconnected environments.
  • DocumentDB is in preview on Azure Local: Support document-oriented applications with local database capabilities.
  • Blob Storage is in preview on Azure Local: Store and access unstructured data with cloud-native object storage while keeping data local.

Together, these services enable organizations to build data-driven and AI-powered applications while keeping sensitive information within their environments.

Sovereign developer workflows with GitHub Enterprise Local

GitHub Enterprise Local, now generally availability, brings GitHub Enterprise Server into Azure Local as a prebuilt virtual appliance. Repositories, CI/CD workflows, and development artifacts stay within the organization’s infrastructure and security boundaries, in both connected and disconnected environments.

In connected environments, developers can use GitHub Copilot with locally hosted repositories. In disconnected environments, Foundry Local enables AI-powered development using locally hosted models.

This gives organizations flexibility to develop and manage applications across connectivity scenarios while maintaining control over their code and infrastructure.

Read more: GitHub Enterprise Local: Bringing modern development to environments you control.

Multitenancy on Azure Local

Multitenancy support on Azure Local is coming soon, giving enterprises and service providers more flexibility in how they use and allocate shared infrastructure. Through logical isolation, organizations can consolidate workloads and use capacity more efficiently while maintaining separation across tenants, teams, and workloads. Initial access will be limited to qualified customers. Sign up for the upcoming preview here.

Unified and secured migration to Azure Local

Azure Local migrations are now part of the modern, unified Azure Catalyst experience, so customers can discover, replicate, and track VMware and Hyper-V migrations all in one place. By the end of the year, we'll have a preview of a new data plane for VMware migrations with Azure Catalyst, formerly known as Azure Migrate to Azure Local, eliminating the dependency on VMware Virtual Disk Development Kit components for supported migration scenarios.

Built on VMware SCSI HotAdd, the new data plane will let customers continue using familiar Azure Catalyst workflows while reducing setup and operational complexity. Existing VDDK-based migration projects remain supported.

Azure Catalyst enables customers to preserve existing Azure Arc-enabled server resources when migrating supported VMware and Hyper-V workloads to Azure Local, retaining RBAC assignments, policies, tags, monitoring configurations, and extensions.

We're expanding Azure Catalyst with private endpoint support, enabling migration-related communication with Azure resources over private connections. General availability is targeted for Q1 CY2027.

And we're announcing general availability of Trusted Launch VM migration, enabling eligible workloads to migrate directly into Trusted Launch configurations on Azure Local.

Sovereign Private Cloud ecosystem

A strong partner ecosystem is essential to helping customers deploy and scale Sovereign Private Cloud. At Ignite, we're expanding partner expertise and application validation to accelerate adoption.

We're launching the Azure Local for Hybrid and Sovereign Private Cloud specialization in the Microsoft Cloud Partner Program. This specialization recognizes partners with proven expertise in designing, deploying, and operating Azure Local solutions. The Hybrid and Sovereign Private Cloud Specialization currently includes ten partners: abtis, Acuutech, Dell, HCLTech, Impresoft 4ward, kloud klickers, Lenovo, NTT DATA, PulsivIT, and Twoday.

We're also working with independent software vendors and global systems integrators to broaden the portfolio of applications validated for Azure Local, spanning enterprise workloads, data platforms, and sovereign AI.

Together, these investments help customers move from initial deployments to production with greater confidence.

Explore the full list of validated solutions here.

The innovations announced at Microsoft Frontier Tour and Ignite build on our Sovereign Private Cloud vision, extending Azure Local with new capabilities across AI, operational flexibility, confidential computing, data services, and developer workflows.  

Our vision remains simple: bring the power of Azure wherever customers need it, with the flexibility and control to operate on their terms.

Learn more at Microsoft Ignite

 

 

Updated Nov 17, 2026
Version 2.0
No CommentsBe the first to comment