Adding Create a Storage Spaces Direct Campus Cluster article

Author: robinharwood

WindowsServerDocs/failover-clustering/create-storage-spaces-direct-campus-cluster.md

@@ -0,0 +1,275 @@
+---
+title: Create a Storage Spaces Direct Campus Cluster on Windows Server
+description: Discover step-by-step instructions to deploy a Storage Spaces Direct campus cluster for high availability and fault tolerance on Windows Server 2025.
+#customer intent: As an IT administrator, I want to create a Storage Spaces Direct campus cluster so that I can ensure rack-level resiliency for my workloads across two physical locations.
+author: robinharwood
+ms.author: roharwoo
+ms.reviewer: robhind
+ms.date: 01/20/2026
+ms.topic: how-to
+---
+
+# Create a Storage Spaces Direct campus cluster
+
+Learn how to create a Storage Spaces Direct campus cluster on Windows Server 2025 and later. A campus cluster distributes data across two physical racks in separate rooms or buildings within the same campus, providing rack-level resiliency for your workloads.
+
+> [!NOTE]
+> A campus cluster is different from a stretch cluster. Campus clusters use LAN connectivity within the same location with S2D replication, while stretch clusters span geographically distant sites connected by WAN and use Storage Replica for replication.
+
+For more information about campus cluster architecture, supported configurations, and how it compares to other topologies, see [Failover Clustering topologies](topologies.md).
+
+## Prerequisites
+
+Before you create a campus cluster, ensure you meet the following requirements:
+
+- **Windows Server 2025** with the 2025-12 Security Update (KB5072033) installed on every node in the failover cluster.
+- **Flat S2D storage** - All capacity drives must be flash-based (SSD or NVMe). Don't use HDDs.
+- **Exactly two RACK fault domains** - You must define exactly two rack-level fault domains and place the cluster nodes in these two racks.
+- **Hardware OEM guidelines** - Follow your hardware OEM's guidelines for the deployment.
+- **Cluster quorum resource** - Place the quorum resource (File Share Witness, Disk Witness, Cloud Witness, or USB Witness) in a third room, separate from the data rooms containing the racks.
+
+For optimal performance and resiliency, use the following configurations:
+
+- Each rack has a separate network path to the cluster quorum resource.
+- Use redundant top-of-rack (TOR) switches, core switches, and dedicated networks for S2D storage traffic to minimize single points of failure and maximize workload uptime and durability.
+- Network latency of 1 ms or less between racks. Use the [PsPing](/sysinternals/downloads/psping) utility to measure network latency.
+- RDMA NICs and switches, which can achieve up to 30% CPU savings.
+
+## Create the failover cluster
+
+To create a campus cluster, first create a failover cluster without storage, then define fault domains before enabling Storage Spaces Direct.
+
+#### [PowerShell](#tab/powershell)
+
+Here's how to create the failover cluster using the [New-Cluster](/powershell/module/failoverclusters/new-cluster) cmdlet.
+
+1. Run PowerShell on your computer in an elevated session.
+
+1. Run the following command to create the cluster without storage:
+
+   ```powershell
+   New-Cluster -Name <ClusterName> -Node <Node1,Node2,Node3,Node4> -NoStorage
+   ```
+
+   Replace `<ClusterName>` with your cluster name and `<Node1,Node2,Node3,Node4>` with your node names.
+
+#### [Failover Cluster Manager](#tab/fcm)
+
+Here's how to create the failover cluster using Failover Cluster Manager.
+
+1. From the Windows desktop, open the **Start** menu, and then select the **Server Manager** tile.
+
+1. On the **Tools** menu, select **Failover Cluster Manager**.
+
+1. In the **Failover Cluster Manager** pane, under **Management**, select **Create Cluster**. The Create Cluster Wizard opens.
+
+1. On the **Before You Begin** page, select **Next**.
+
+1. On the **Select Servers** page, in the **Enter name** box, enter the name of a server that you plan to add as a failover cluster node, and then select **Add**. Repeat this step for each of the four nodes. When you're finished, select **Next**.
+
+1. On the **Validation Warning** page, select **Yes** to run cluster validation, and then select **Next**. Complete the Validate a Configuration Wizard.
+
+1. On the **Access Point for Administering the Cluster** page, enter a name for the cluster. If the servers don't use DHCP, configure one or more static IP addresses for the failover cluster. Select **Next**.
+
+1. On the **Confirmation** page, clear the **Add all eligible storage to the cluster** checkbox because you configure storage later when enabling Storage Spaces Direct. Select **Next**.
+
+1. On the **Summary** page, confirm that the failover cluster was created successfully, and then select **Finish**.
+
+---
+
+## Define fault domains
+
+After you create the cluster, define fault domains before you enable Storage Spaces Direct.
+
+> [!IMPORTANT]
+> You must define fault domains before enabling Storage Spaces Direct. Once you create the pool and volumes, data doesn't retroactively move in response to changes to the fault domain topology.
+
+#### [PowerShell](#tab/powershell)
+
+Here's how to define fault domains using PowerShell.
+
+1. Run PowerShell on your computer in an elevated session.
+
+1. Run the following command to define two rack fault domains, with two nodes in each rack:
+
+   ```powershell
+   Set-ClusterFaultDomain -XML @"
+   <Topology>
+     <Site Name="Campus">
+       <Rack Name="Room1">
+         <Node Name="Node1"/>
+         <Node Name="Node2"/>
+       </Rack>
+       <Rack Name="Room2">
+         <Node Name="Node3"/>
+         <Node Name="Node4"/>
+       </Rack>
+     </Site>
+   </Topology>
+   "@
+   ```
+
+   Replace the node names with your actual node names, and update the site and rack names as appropriate for your environment.
+
+1. Verify the fault domain configuration by running:
+
+   ```powershell
+   Get-ClusterFaultDomain
+   ```
+
+For alternative methods to define fault domains, including using individual PowerShell cmdlets, see [Fault domain awareness](fault-domains.md).
+
+#### [Failover Cluster Manager](#tab/fcm)
+
+Failover Cluster Manager doesn't support configuring rack-level fault domains through the GUI. Use PowerShell to define fault domains.
+
+1. Run PowerShell on your computer in an elevated session.
+
+1. Run the following command to define two rack fault domains, with two nodes in each rack:
+
+   ```powershell
+   Set-ClusterFaultDomain -XML @"
+   <Topology>
+     <Site Name="Campus">
+       <Rack Name="Room1">
+         <Node Name="Node1"/>
+         <Node Name="Node2"/>
+       </Rack>
+       <Rack Name="Room2">
+         <Node Name="Node3"/>
+         <Node Name="Node4"/>
+       </Rack>
+     </Site>
+   </Topology>
+   "@
+   ```
+
+   Replace the node names with your actual node names.
+
+1. Verify the fault domain configuration by running:
+
+   ```powershell
+   Get-ClusterFaultDomain
+   ```
+
+---
+
+## Enable Storage Spaces Direct
+
+After defining fault domains, enable Storage Spaces Direct on the cluster.
+
+#### [PowerShell](#tab/powershell)
+
+Here's how to enable Storage Spaces Direct using the [Enable-ClusterStorageSpacesDirect](/powershell/module/failoverclusters/enable-clusterstoragespacesdirect) cmdlet.
+
+1. Run PowerShell on your computer in an elevated session.
+
+1. Enable Storage Spaces Direct on the cluster:
+
+   ```powershell
+   Enable-ClusterStorageSpacesDirect
+   ```
+
+1. Update the storage pool to ensure it's using the latest features:
+
+   ```powershell
+   Get-StoragePool S2D* | Update-StoragePool
+   ```
+
+1. Verify that the storage pool version is 29 or later:
+
+   ```powershell
+   (Get-CimInstance -Namespace root/microsoft/windows/storage -ClassName MSFT_StoragePool -Filter 'IsPrimordial = false').CimInstanceProperties['Version'].Value
+   ```
+
+1. Verify that the storage pool's `FaultDomainAwareness` property is set to `StorageRack`:
+
+   ```powershell
+   Get-StoragePool -FriendlyName S2D* | Format-List FaultDomainAwarenessDefault
+   ```
+
+#### [Failover Cluster Manager](#tab/fcm)
+
+Here's how to enable Storage Spaces Direct using Failover Cluster Manager.
+
+1. In **Failover Cluster Manager**, expand the cluster name, and then select **Storage** > **Pools**.
+
+1. Right-click **Pools**, and then select **Enable Storage Spaces Direct**. Follow the wizard to complete the operation.
+
+1. After enabling Storage Spaces Direct, use PowerShell to update the storage pool and verify the configuration:
+
+   ```powershell
+   Get-StoragePool S2D* | Update-StoragePool
+   ```
+
+1. Verify that the storage pool version is 29 or later:
+
+   ```powershell
+   (Get-CimInstance -Namespace root/microsoft/windows/storage -ClassName MSFT_StoragePool -Filter 'IsPrimordial = false').CimInstanceProperties['Version'].Value
+   ```
+
+1. Verify that the storage pool's `FaultDomainAwareness` property is set to `StorageRack`:
+
+   ```powershell
+   Get-StoragePool -FriendlyName S2D* | Format-List FaultDomainAwarenessDefault
+   ```
+
+---
+
+## Create volumes
+
+Create volumes on the storage pool based on your resiliency requirements. For a 2+2 campus cluster, use four-copy volumes as they provide the best balance between cost, performance, and resiliency. A four-copy volume places a copy of data on each node, so the cluster can survive losing an entire rack and a node at the same time.
+
+A four-copy volume provides 25% storage capacity, while a two-copy volume provides 50% capacity with less resiliency. For more information about volume resiliency options, see [Fault tolerance and storage efficiency](../storage/storage-spaces/fault-tolerance.md).
+
+#### [PowerShell](#tab/powershell)
+
+Here's how to create a four-copy mirrored volume using PowerShell.
+
+1. Run PowerShell on your computer in an elevated session.
+
+1. Run the following command to create a four-copy volume:
+
+   ```powershell
+   New-Volume -FriendlyName "FourCopyVolume" `
+       -StoragePoolFriendlyName S2D* `
+       -FileSystem CSVFS_ReFS `
+       -Size 500GB `
+       -ResiliencySettingName Mirror `
+       -PhysicalDiskRedundancy 3 `
+       -ProvisioningType Fixed `
+       -NumberOfDataCopies 4 `
+       -NumberOfColumns 3
+   ```
+
+   To create a two-copy volume instead, set `-PhysicalDiskRedundancy 1` and remove the `-NumberOfDataCopies` and `-NumberOfColumns` parameters. To use thin provisioning, change `-ProvisioningType` to `Thin`.
+
+For more information about resiliency types and storage efficiency, see [Fault tolerance and storage efficiency](../storage/storage-spaces/fault-tolerance.md).
+
+#### [Failover Cluster Manager](#tab/fcm)
+
+Here's how to create volumes using Failover Cluster Manager. To create four-copy volumes, use PowerShell.
+
+1. In **Failover Cluster Manager**, expand the cluster name, and then select **Storage** > **Pools**.
+
+1. Select the **S2D** storage pool, and then right-click and select **New Virtual Disk**.
+
+1. In the New Virtual Disk Wizard, enter a name for the volume.
+
+1. On the **Storage Layout** page, select **Mirror**.
+
+1. On the **Resiliency Settings** page, select **Two-way mirror** for a two-copy volume.
+
+1. On the **Provisioning Type** page, select either **Fixed** or **Thin**.
+
+1. Specify the size for the volume and complete the wizard.
+
+1. After creating the virtual disk, right-click it and select **New Volume** to format it by using ReFS.
+
+---
+
+## Related content
+
+- [Deploy a quorum witness](deploy-quorum-witness.md)
+- [Fault domain awareness](fault-domains.md)
+- [Announcing support for S2D Campus Cluster on Windows Server 2025](https://techcommunity.microsoft.com/blog/failoverclustering/announcing-support-for-s2d-campus-cluster-on-windows-server-2025/4477075)

WindowsServerDocs/failover-clustering/toc.yml

@@ -8,6 +8,8 @@
   items:
   - name: Create a failover cluster
     href: create-failover-cluster.md
+  - name: Create a Storage Spaces Direct campus cluster
+    href: create-storage-spaces-direct-campus-cluster.md
   - name: Deploy a two-node file server
     href: deploy-two-node-clustered-file-server.md
   - name: Recover a failover cluster without quorum