System requirements for Azure Stack HCI
Applies to: Azure Stack HCI, version 20H2
Use this topic to assess system requirements for servers, storage, and networking for Azure Stack HCI.
A standard Azure Stack HCI cluster requires a minimum of two servers and a maximum of 16 servers; however, clusters can be combined using cluster sets to create an HCI platform of hundreds of nodes.
Keep the following in mind for various types of Azure Stack HCI deployments:
Stretched clusters require servers to be deployed at two separate sites. The sites can be in different countries, different cities, different floors, or different rooms. A stretched cluster requires a minimum of 4 servers (2 per site) and a maximum of 16 servers (8 per site).
It's recommended that all servers be the same manufacturer and model, using 64-bit Intel Nehalem grade, AMD EPYC grade, or later compatible processors with second-level address translation (SLAT). A second-generation Intel Xeon Scalable processor is required to support Intel Optane DC persistent memory. Processors must be at least 1.4 GHz and compatible with the x64 instruction set.
Make sure that the servers are equipped with at least 32 GB of RAM per node to accommodate the server operating system, VMs, and other apps or workloads. In addition, allow 4 GB of RAM per terabyte (TB) of cache drive capacity on each server for Storage Spaces Direct metadata.
Verify that virtualization support is turned on in the BIOS or UEFI:
- Hardware-assisted virtualization. This is available in processors that include a virtualization option, specifically processors with Intel Virtualization Technology (Intel VT) or AMD Virtualization (AMD-V) technology.
- Hardware-enforced Data Execution Prevention (DEP) must be available and enabled. For Intel systems, this is the XD bit (execute disable bit). For AMD systems, this is the NX bit (no execute bit).
You can use any boot device supported by Windows Server, which now includes SATADOM. RAID 1 mirror is not required, but is supported for boot. A 200 GB minimum size is recommended.
For additional feature-specific requirements for Hyper-V, see System requirements for Hyper-V on Windows Server.
Azure Stack HCI works with direct-attached SATA, SAS, NVMe, or persistent memory drives that are physically attached to just one server each.
For best results, adhere to the following:
Every server in the cluster should have the same types of drives and the same number of each type. It's also recommended (but not required) that the drives be the same size and model. Drives can be internal to the server, or in an external enclosure that is connected to just one server. To learn more, see Drive symmetry considerations.
Each server in the cluster should have dedicated volumes for logs, with log storage at least as fast as data storage. Stretched clusters require at least two volumes: one for replicated data, and one for log data.
SCSI Enclosure Services (SES) is required for slot mapping and identification. Each external enclosure must present a unique identifier (Unique ID).
NOT SUPPORTED: RAID controller cards or SAN (Fibre Channel, iSCSI, FCoE) storage, shared SAS enclosures connected to multiple servers, or any form of multi-path IO (MPIO) where drives are accessible by multiple paths. Host-bus adapter (HBA) cards must implement simple pass-through mode.
An Azure Stack HCI cluster requires a reliable high-bandwidth, low-latency network connection between each server node.
- Verify at least one network adapter is available and dedicated for cluster management.
- Verify that physical switches in your network are configured to allow traffic on any VLANs you will use.
For host networking considerations and requirements, see Host network requirements.
Software Defined Networking (SDN) requirements
When you create an Azure Stack HCI cluster using Windows Admin Center, you have the option to deploy Network Controller to enable Software Defined Networking (SDN). If you intend to use SDN on Azure Stack HCI:
Make sure the host servers have at least 50-100 GB of free space to create the Network Controller VMs.
You must copy a virtual hard disk (VHD) of the Azure Stack HCI operating system to the first node in the cluster in order to create the Network Controller VMs. You can prepare the VHD using Sysprep or by running the Convert-WindowsImage script to convert an .iso file into a VHD.
SDN is not supported on stretched (multi-site) clusters.
SDN hardware requirements
This section provides network hardware requirements for NICs and physical switches when planning for a SDN environment.
Network interface cards (NICs)
The NICs that you use in your Hyper-V hosts and storage hosts require specific capabilities to achieve the best performance.
Remote Direct Memory Access (RDMA) is a kernel bypass technique that makes it possible to transfer large amounts of data without using the host CPU, which frees the CPU to perform other work. Switch Embedded Teaming (SET) is an alternative NIC Teaming solution that you can use in environments that include Hyper-V and the SDN stack. SET integrates some NIC Teaming functionality into the Hyper-V Virtual Switch.
For more information, see Remote Direct Memory Access (RDMA) and Switch Embedded Teaming (SET).
To account for the overhead in tenant virtual network traffic caused by VXLAN or NVGRE encapsulation headers, the maximum transmission unit (MTU) of the Layer-2 fabric network (switches and hosts) must be set to greater than or equal to 1674 bytes (including Layer-2 Ethernet headers).
NICs that support the new EncapOverhead advanced adapter keyword set the MTU automatically through the Network Controller Host Agent. NICs that do not support the new EncapOverhead keyword need to set the MTU size manually on each physical host using the JumboPacket (or equivalent) keyword.
Switches and routers
When selecting a physical switch and router for your SDN environment, make sure it supports the following set of capabilities:
- Switchport MTU settings (required)
- MTU set to >= 1674 bytes (including L2-Ethernet Header)
- L3 protocols (required)
- Equal-cost multi-path (ECMP) routing
- BGP (IETF RFC 4271)-based ECMP
Implementations should support the MUST statements in the following IETF standards:
- RFC 2545: BGP-4 Multiprotocol extensions for IPv6 Inter-Domain Routing
- RFC 4760: Multiprotocol Extensions for BGP-4
- RFC 4893: BGP Support for Four-octet AS Number Space
- RFC 4456: BGP Route Reflection: An Alternative to Full Mesh Internal BGP (IBGP)
- RFC 4724: Graceful Restart Mechanism for BGP
The following tagging protocols are required:
- VLAN - Isolation of various types of traffic
- 802.1q trunk
The following items provide Link control:
- Quality of Service (QoS) (PFC only required if using RoCE)
- Enhanced Traffic Selection (802.1Qaz)
- Priority-based Flow Control (PFC) (802.1p/Q and 802.1Qbb)
The following items provide availability and redundancy:
- Switch availability (required)
- A highly available router is required to perform gateway functions. You can provide this by using either a multi-chassis switch\router or technologies like the Virtual Router Redundancy Protocol (VRRP).
There are no special domain functional level requirements for Azure Stack HCI - just an operating system version for your domain controller that's still supported. We do recommend turning on the Active Directory Recycle Bin feature as a general best practice, if you haven't already. to learn more, see Active Directory Domain Services Overview.
Maximum supported hardware specifications
Azure Stack HCI deployments that exceed the following specifications are not supported:
|Physical servers per cluster||16|
|VMs per host||1,024|
|Disks per VM (SCSI)||256|
|Storage per cluster||4 PB|
|Storage per server||400 TB|
|Logical processors per host||512|
|RAM per host||24 TB|
|RAM per VM||12 TB (generation 2 VM) or 1 TB (generation 1)|
|Virtual processors per host||2,048|
|Virtual processors per VM||240 (generation 2 VM) or 64 (generation 1)|
For related information, see also: