Loading...

GlusterFS Subclients

GlusterFS supports the following types of subclients:

  • Default subclients
  • User-defined subclients

When you create an instance, a default subclient is automatically created. You can configure only one GlusterFS volume for a subclient. You can create user-defined subclients to protect different groups of data.

You can change the data access nodes that are associated with the default subclient and with the user-defined subclients.

GlusterFS Scan Optimization using GlusterFind API

GlusterFS scan performance is improved by using the GlusterFind API.

Commvault software runs the scan operation on the GlusterFS clients, but the GlusterFind query runs on the GlusterFS cluster. The GlusterFind query results are staged on the GlusterFS cluster node in the /tmp directory by default. Then, the query results are moved to the JobResults directory on the data access node that runs the scan operation and processed on that node. You can configure a different path for the GlusterFind query results, see Configuring a Different Path for the GlusterFind Query Results.

Performance Comparison

The following table compares performance metrics between GlusterFind scan and regular scan on Linux.

Job Type

Number of Files

GlusterFind Scan Time

Regular Scan Time

Incremental backup job without TrueUp

106,695 new files added

01:43:12

29:30:11

Synchronizing Data on the Disk and the Index for GlusterFS

Synchronizing data on the disk and the index operation runs GlsuterFind full scans and might run longer.

Last modified: 8/8/2018 11:35:09 AM