Monday, November 10, 2008

SharePoint Backup Models

Types and Descriptions of Backup Models as suggested by Microsoft.

"Full Backup Model w/ Truncation
The full recovery model uses log backups to prevent data loss in the broadest range of failure scenarios, and backing and restoring the transaction log (log backups) is required. The advantage of using log backups is that they let you to restore a database to any point of time that is contained within a log backup (point-in-time recovery). Assuming you can back up the active log after a disaster occurs; you can restore the database up to the point of failure without data loss. The disadvantages of using log backups are that they require storage space and increase restore time and complexity.

Truncation
Under the full recovery model or bulk-logged recovery model, the inactive part of the log cannot be truncated until all its log records have been captured in a log backup. This is needed to maintain the log chain—a series of log records having an unbroken sequence of log sequence numbers (LSNs). The log is truncated when you back up the transaction log.

Optional Backup Models

Simple Backup Model
The simple recovery model provides the simplest form of backup and restore. Backup is easy to manage because the transaction log is never backed up. However, if there are no log backups, a database can be restored only to the end of the most recent backup of the data. If a failure were to occur, updates that are made after the most recent backup of the data are lost.

Bulk Logged Backup Model
This topic is relevant for optimizing bulk operations on SQL Server databases that typically use the full recovery model.

The bulk-logged recovery model is a special-purpose recovery model that should be used only intermittently to improve the performance of certain large-scale bulk operations, such as bulk imports of large amounts of data. Much of the description of backup under the full recovery model also applies to the bulk-logged recovery model. This topic looks only at considerations that are unique to the bulk-logged recovery model." -Microsoft

-Brian Grabowski

No comments: