Administrator Guide

Synchronous replication use cases
25 Dell EMC SC Series: Synchronous Replication and Live Volume | CML1064
high availability mode recovery, disaster recovery, or remote replicas which will be discussed in the coming
sections.
High availability synchronous with databases
A replication link or destination volume issue in St. Paul results in no database outage in
Minneapolis
4.4 Remote database replicas
One practice commonly found in organizations with Microsoft SQL Server or Oracle database technologies is
to create copies of databases. There are several reasons to clone databases, and most of them stem from
the common principle of minimal or no disruption to the production database, and thus to the application and
end users. To identify a few examples, at least one separate copy of a database is maintained for application
developers to develop and test code against. A separate database copy is maintained for DBA staff to test
index changes, queries, and for troubleshooting areas such as performance. A copy of the production
database may be maintained for I/O intensive queries or reporting. SC Series storage snapshots and View
Volumes are a natural tactical fit for fulfilling database replica needs locally on the same SC Series array.
However, if the replica is to be stored on a different array, whether or not it is in the same building or
geographic region, replication or portable volume must be used to seed the data remotely, and replication
should be used to refresh the data as needed. For the purposes of developer or DBA testing, asynchronous
replication may be timely enough. However, for reporting purposes, synchronous replication will ensure the
data is up to date when the reporting database is refreshed using the replicated volumes. The choice of
providing zero data loss through high consistency mode or a more flexible high availability mode should be
decided ahead of time with the impacts of each mode well understood.
SC Series snapshots, as well as asynchronous and synchronous replication, are natively space and
bandwidth efficient on storage and replication links respectively. Only the changed data is frozen in a
snapshot and replicated to remote SC Series arrays. In the following figure, notice the use of high availability
synchronous replication within the Minneapolis data center. Although the two arrays are well connected, the
risk of internal reporting database inconsistency does not warrant a production outage for the organization.