Blog Post

Azure Arc Blog
3 MIN READ

Announcing the preview of Azure Local rack aware cluster

mindydiep's avatar
mindydiep
Icon for Microsoft rankMicrosoft
Nov 18, 2025

We are excited to announce the public preview of Azure Local rack aware cluster! We previously published a blog post with a sneak peek of Azure Local rack aware cluster and now, we're excited to share more details about its architecture, features, and benefits.  

Overview of Azure Local rack aware cluster 

Azure Local rack aware cluster is an advanced architecture designed to enhance fault tolerance and data distribution within an Azure Local instance. This solution enables you to cluster machines that are strategically placed across two physical racks in different rooms or buildings, connected by high bandwidth and low latency within the same location. Each rack functions as a local availability zone, spanning layers from the operating system to Azure Local management, including Azure Local VMs.

The architecture leverages top-of-rack (ToR) switches to connect machines between rooms. This direct connection supports a single storage pool, with rack aware clusters distributing data copies evenly between the two racks. Even if an entire rack encounters an issue, the other rack maintains the integrity and accessibility of the data. This design is valuable for environments needing high availability, particularly where it is essential to avoid rack-level data loss or downtime from failures like fires or power outages.

 

Network architecture diagram of an Azure Local rack aware cluster instance

Key features 

Starting in Azure Local version 2510, this release includes the following key features for rack aware clusters: 

Rack-Level Fault Tolerance & High Availability  

Clusters span two physical racks in separate rooms, connected by high bandwidth and low latency. Each rack acts as a local availability zone. If one rack fails, the other maintains data integrity and accessibility. 

Support for Multiple Configurations  

Architecture supports 2 machines up to 8 machines, enabling scalable deployments for a wide range of workloads. 

Scale-Out by Adding Machines  

Easily expand cluster capacity by adding machines, supporting growth and dynamic workload requirements without redeployment. 

Unified Storage Pool with Even Data Distribution 

Rack aware clusters offer a unified storage pool with Storage Spaces Direct (S2D) volume replication, automatically distributing data copies evenly across both racks. This ensures smooth failover and reduces the risk of data loss. 

Azure Arc Integration and Management Experience 

Enjoy native integration with Azure Arc, enabling consistent management and monitoring across hybrid environments—including Azure Local VMs and AKS—while maintaining the familiar Azure deployment and operational experience. 

Deployment Options 

Deploy via Azure portal or ARM templates, with new inputs and properties in the Azure portal for rack aware clusters. 

Screenshot of deployment wizard in Azure portal for Azure Local rack aware cluster

Provision VMs in Local Availability Zones via the Azure Portal 

Provision Azure Local virtual machines directly into specific local availability zones using the Azure portal, allowing for granular workload placement and enhanced resilience. 

Screenshot of Azure Arc virtual machine creation wizard in Azure portal

Upgrade Path from Preview to GA 

Deploy rack aware clusters with the 2510 public preview build and update to General Availability (GA) without redeployment—protecting your investment and ensuring operational continuity. 

Get started 

The preview of rack aware cluster is now available to all interested customers. We encourage you to try it out and share your valuable feedback. To get started, visit our documentation: Overview of Azure Local rack aware clustering (Preview) - Azure Local | Microsoft Learn 

Stay tuned for more updates as we work towards general availability in 2026. We look forward to seeing how you leverage Azure Local rack aware cluster to power your edge workloads!

Updated Nov 12, 2025
Version 1.0
No CommentsBe the first to comment