Home > Storage > PowerFlex > White Papers > Dell APEX Block Storage for Public Cloud: Microsoft SQL Server Multi-Subnet Failover Clustering Deployment > Logical architecture
The logical architecture described in this section shows the deployment of a three node SQL Server 2022 multisubnet failover cluster on Windows Server 2022 on a two-layer Dell APEX Block Storage configuration - SQL Server 2022 Node 1 in AZ-1a, SQL Server 2022 Node 2 in AZ-1b, and SQL Server 2022 Node 3 in AZ-1c. SQL Server setup classifies this configuration as a multisubnet cluster. The APEX Block Storage deployment across three AZs offers increased resiliency and availability.
Figure 5 represents the two-layer APEX Block Storage system using the Amazon instance store (natively attached to NVMe SSD drives).
In each AZ, one SDS is installed on i3en.12xlarge Amazon EC2 instance with the Red Hat Enterprise Linux AMI. The Amazon EC2 instances that are deployed by APEX Block Storage in AWS are backed by NVMe SSD drives. The NVMe SSDs are locally attached at the Instance level. The natively attached NVMe SSD drives are ephemeral storage. There are three APEX Block Storage SDS instances that are configured into a single protection domain. A single storage pool is created, using all the storage devices available within the protection domain.
APEX Block Storage maintains the user data in a mesh mirrored layout such that each piece of data has two copies (primary and secondary) stored on two different storage optimized Amazon EC2 instances (SDS). The copies are evenly distributed across the SDS instances, which enables the system to maintain data availability and high-performance if there is a failure of a storage device or Amazon EC2 storage optimized instance.
The APEX Block Storage volumes that are created from the APEX Block Storage pool are mapped to Amazon EC2 compute instances running APEX Block Storage SDC services. The APEX Block Storage volumes mapped to instances are used to create the file system to host the databases.