Azure Stack HCI is a robust solution for businesses seeking a hyper-converged infrastructure (HCI) platform. To set up a standard Azure Stack HCI cluster, it’s essential to meet specific hardware requirements. In this post, we’ll delve into the key considerations and prerequisites for deploying Azure Stack HCI effectively.
- Cluster Size and Configuration
A standard Azure Stack HCI cluster typically consists of a minimum of one server and can scale up to a maximum of 16 servers. However, through the utilization of cluster sets, it’s possible to create an HCI platform that incorporates hundreds of nodes.
- Networking
Azure Stack HCI clusters can be configured as RDMA Direct (Switchless) up to 5 node servers within a single cluster, this allows the capability to re-use existing Top of Rack (ToR) switches without the need for a costly network refresh.
- Stretched Clusters
For stretched clusters, servers need to be deployed across two separate sites, which could be in different countries, cities, floors, or rooms. Stretched clusters require a minimum of four servers (two at each site) and can extend to a maximum of 16 servers (eight at each site).
- Server Specifications
Ensure that all servers are of the same manufacturer and model, equipped with 64-bit Intel Nehalem grade, AMD EPYC grade, or later compatible processors featuring second-level address translation (SLAT). Furthermore, second-generation, or later, Intel Xeon Scalable processors are essential to support Intel Optane DC persistent memory. Processors should operate at a minimum of 1.4 GHz and be compatible with the x64 instruction set.
- Memory Requirements
Each server node should be equipped with a minimum of 64 GB of RAM. This allocation is necessary to accommodate the server operating system, virtual machines (VMs), and other applications or workloads. Additionally, allocate an extra 4 GB of RAM per terabyte (TB) of cache drive capacity on each server for Storage Spaces Direct metadata.
- Virtualization Support
In the BIOS or UEFI settings of your servers, verify that virtualization support is enabled. This involves two key components:
- Hardware-assisted virtualization is available in processors with options like Intel Virtualization Technology (Intel VT) or AMD Virtualization (AMD-V) technology.
- Hardware-enforced Data Execution Prevention (DEP), which should be available and enabled. On Intel systems, this is referred to as the XD bit (execute disable bit), while on AMD systems, it’s known as the NX bit (no execute bit).
- Boot Device
Azure Stack HCI supports a variety of boot devices, including SATADOM. While a RAID 1 mirror is not mandatory, it is a supported configuration for boot purposes. It’s advisable to use a boot device with a minimum size of 200 GB for optimal performance.
- Windows Server Hot Patching
As of December 2021, Windows Server Hot Patching is available for on-premises Azure Stack HCI deployments. This feature enhances system maintenance and minimizes downtime.
For specific requirements related to Hyper-V and additional details on logical and system requirements, please refer to the Microsoft Documentation site dedicated to Azure Stack HCI.
By adhering to these hardware prerequisites, you can ensure a smooth and efficient deployment of Azure Stack HCI, empowering your organization with a powerful hyper-converged infrastructure solution.
Get in Touch!
Are you ready to supercharge your hybrid cloud strategy? We’re here to help. Email us at sales@hcistacksservices.com or tony@hcistacksservices.com to discuss how this powerful technology can benefit your business.