Tuesday, November 29, 2011

Defrag.NSF features - Part 5

Continuing with the Technical Settings options we look at the ability to schedule compacts and the various selection methods.

Run Compact before scheduled defrag with option
With this option enabled Defrag.NSF will run the Compact server task with the configured Compact switch on each processed database prior to the actual defragmentation of the database. This is a handy option and can be used instead of a Program Document to run weekly scheduled Compact maintenance automatically followed by a defrag of the file to clean-up any fragmentation

introduced by the compact process. This is a perfect scenario just prior to running backup maintenance as the files will be at their smallest size and also freshly defragged providing maximum file read performance and data throughput for your backup software.


Selection Method
Defrag.NSF uses the selection methods configured here as “rules” for identifying which databases will be tagged for defragging during the next maintenance run. When running on a schedule Defrag will perform the Cataloging process first, during this Cataloging process Defrag.NSF checks all the databases against these Selection Methods and tags the databases that fall within the configured rules. Defrag.NSF will then perform the defragging of any files tagged via the Selection Method. If any server maintenance tasks such as Compact or Fixup were scheduled using the Technical Settings options, these tasks will be performed immediately prior to the defrag operation.

Option – When fragments exceed [n]

Use this option when you want to target for defragmentation all databases exceeding this configured number of fragments.

Option – When fragmentation Health Index exceeds [n]

Use this option when you want to target for defragmentation all databases exceeding a configured Health Index. For a detailed explanation of the Health Index, please read the previous section on the Health Index view.

Option – The top databases [n]

Use this option when you just want to always target the most fragmented databases on the server. This option uses fragment count only as the criteria for selecting the database for maintenance.

Next we will look at the scheduling options offered in Defrag.NSF

No comments: