White Papers
11 TR1081 | Microsoft SQL Server Database Protection Using EqualLogic Auto-Snapshot Manager / Microsoft Edition
4 Setup and configuration best practices
This section details some best practices for using ASM/ME with SQL Server and PS Series groups.
4.1 Database File Layout
To create SQL Server database Smart Copies with ASM/ME, the SQL Server host must be on the same
iSCSI network as the PS Series Group and all of the database files must reside on PS Series volumes
presented through the SQL server host initiator. ASM/ME recognizes all the underlying volumes that make
up the database object.
PS Series Groups enable volume-based data protection. All the data on a selected volume will be
protected during a Smart Copy operation. ASM/ME creates an application-consistent Smart Copy of the
object or database chosen even if separate database files reside on the same volume. If more than one
database shares the same volume, it could result in a ”torn” Smart Copy set of the database(s) not chosen.
To avoid torn smart copies ASM/ME will recognize and prompt the user to select the other objects or
databases that reside on the same volumes.
Best Practices: In environments where volumes share database files with other databases, it’s
recommended to select all the databases sharing similar volumes to avoid torn Smart Copies.
Note: During restores in these environments a
Restore All
will roll back all the databases associated with
the volume recovery. Choose a Selective Restore to recover individual databases. See the procedure for
selective recovery of individual databases later in this doc for more information.
4.2 Storage Resource Management
This section describes how ASM/ME and Smart Copies use storage resources and suggests some best
practices for monitoring and managing space used by Smart Copy sets on the PS Series group.
Snapshot Smart Copies use the snapshot reserve space allocated to the volume. The amount of snapshot
reserve can be monitored and changed for each volume that makes up a SQL Server database by viewing
the volume object in ASM/ME and changing reserve allocations in the PS Series group manager.
When a snapshot Smart Copy is restored as a new database (Side-by-Side Restore), the snapshot
continues to exist, and continues to use the snapshot reserve for the volumes that make up the Smart
Copy. Any writes to these snapshots while they are mounted on a host will consume additional snapshot
reserve.
Best Practice: Maintain “keep counts” for Smart Copy schedules to minimize overuse of snapshot reserve.
Side-by-side restores of Snapshot Smart Copies are best used as temporary copies of databases. For long
term use of a Smart Copy it is recommended to use Clone Smart Copies.
Clone Smart Copies are exact copies as the original volume or object. When you create a clone Smart
Copy, a new volume appears in the Group Manager GUI with the date and timestamp of the Smart Copy