HP 9000 Virtual Library System - Manual (Page 85)

entire backup job and to prevent too many backup jobs from piling up on the same cartridge,
but small enough that you are not wasting overall cartridge space.
•
Additional nodes — For systems with long backup windows, you may want to include additional
nodes to speed up the post-processing deduplication. A VLS system using deduplication can
support up to six 1U or 2U high performance nodes.
•
Additional capacity — Because you must store two full backups before the delta-differencing
begins, you may want to include additional disk capacity at least the size of the sum of the
backups. A VLS system using deduplication can support up to eight arrays with four 1U or
2U nodes.
Installing the Deduplication Licenses
The deduplication feature is a licensed option. The number of deduplication licenses must match
or exceed the number of capacity licenses (based on LTU) on the VLS device. See
Installing
Additional Licenses (page 46) for installation instructions.
After installing the required deduplication licenses, the VLS automatically reboots. When the system
is back up, it first checks the standard capacity licenses and then the deduplication licenses. Once
the deduplication licenses are verified, the deduplication feature is automatically enabled for all
supported backup types.
No further setup is required. When the next backup begins, the deduplication process begins
mapping the content of that backup. After the second backup, the system delta-differences
(compares) the two backups and the deduplication proceeds.
You may choose to disable some backup types; see Configuring Deduplication Options (page 85)
for details.
Configuring Deduplication Options
You can include or exclude specific backups or types of backup (e.g., server type) from the
deduplication process and choose the deduplication algorithm used.
NOTE: These options are not available until you have installed the appropriate number of
deduplication licenses.
1. Select the System tab.
2. On the navigation tree, navigate to Chassis
→Deduplication→Configuration .
The Configuration screen displays.
3. For each data type, select the appropriate deduplication algorithm from the list. The defaults
depend on the backup application and the file type; in each case, the algorithm most likely
to result in the best deduplication ratio is used. Depending on your current setting, the options
are:
•
Backup — Useful when file-level differencing is less space efficient (for example, if the
file server is full of very small files).
•
File — Useful for file servers.
•
Enhanced Backup — Useful for databases that are:
Striped ◦
◦ Multiplexed
◦ Large (greater than 14 TB)
◦ Flat-file with deduplication speed or ratio problems
Configuring Deduplication Options 85