Backup of large volumes could benefit of dividing the files into smaller sets and copy them using parallel streams to storage devices. In the case of a tape library, multiple drives can be concurrently used.
In FileSystem plugin options select “Utilise Multiple Streams During Backups”.
Configure the number of streams (8 max) and check the following options as required:
Explanation of these options follows to help implement them correctly.
When backing up a large volume, you can configure the plug-in to dynamically split the data into multiple selection groups and use multiple parallel streams to back up these groups. This task is performed by the Netvault Server, so an increase from the NetVault Server CPU usage is to be expected.
The criteria used to determine the selection groups include the following:
• Number of peer directories
• Number of sub-directories
• Number of files within a directory
After identifying the selection groups, the plug-in creates a child process to back up each selection group. Each child process uses the same criteria to identify parts of its backup that can benefit from a separate stream. If the current stream count has not exceeded the maximum allowed streams, the plug-in assigns a new child process to back up the sub-group. Otherwise, it uses the current stream to back up the sub-group.
During Incremental Backups, only the top-level selection groups use separate data streams; the child processes do not attempt to split the selection groups.
NOTE: This feature is intended for single volume selections that contain large amounts of data (for example, volumes that are larger than 1TB in size). It is not likely to benefit small backups and should not be used in those cases.
When you select this check box, the plug-in uses separate streams to back up the mount points that exist on the volumes. You can use this option with the Use Heuristics for large Single Volume backups option.
© ALL RIGHTS RESERVED. Terms of Use Privacy Cookie Preference Center