White Papers

Overview of Dell EMC SC Series tools for HA/DR
8 SQL Server High Availability and Disaster Recovery with Dell EMC SC Series | CML1119
reports on the amount of data that would be replicated given the selected options. This information can be
used to make better replication architecture decisions. If you are satisfied with the simulated replication, it can
be converted into an actual replication.
1.6 Synchronous Live Volume w/Auto Failover
Always On Failover Cluster Instances (FCIs) are one of the key technologies recommended by Microsoft for
providing High Availability for SQL Server. The recommendation is to use a SAN for the shared cluster
storage. Storage Center has long supported FCIs for local clusters utilizing the same array. However, there
are times when using separate arrays is required. This may be for increased redundancy within the same
data center or stretch cluster configurations within a campus or across separate locations. This is sometimes
referred to as a metro-cluster or geo-cluster. For these scenarios, SC Series Arrays include a feature called
Live Volume that allows the same volume to be presented from two different arrays to the cluster nodes,
fulfilling the shared storage requirement for Always On FCIs.
When using the Live Volume feature with SQL Server, a non-uniform configuration is usually desired,
especially when multiple sites are involved to help ensure that both the primary Live Volume and the SQL
Server instance are running at the same site and front end I/O traffic is located within the site. To support this
configuration, there are some best practices that should be followed. First, Live Volumes should be
synchronous in High Availability mode. High Availability mode allows Auto Failover to be used for the Live
Volume. If the primary volume fails without Auto Failover enabled, Live Volumes will be unavailable until the
secondary is manually recovered. Second, the Swap Roles Automatically option should also be enabled. This
allows the primary and secondary to swap roles so the clustered disk resources are writeable.
It’s important to understand that the SQL Server Failover Cluster Instance is reacting to the loss of storage
visibility to trigger the failover and it is not coordinated in any way by the array or Dell EMC Storage Manager.
Also, there are several configuration options for customizing Live Volume to meet various manual and
automatic failover requirements. The Live Volume with Auto Failover Support for Microsoft video is a great
resource for in-depth understanding of Synchronous Live Volume with Auto Failover.
Note: Replay Manager is not compatible with Live Volume.