Administrator Guide

Table Of Contents
The data reduction settings are shown at the bottom of the status information panel.
NAS Container Data Reduction
Data reduction is a process that runs according to a schedule on each NAS container that has data reduction enabled. A policy that
you dene determines whether or not a le qualies for data reduction, on the basis of access and modication times of that le.
The data reduction process analyzes les to determine if multiple copies of an individual le, or portions of a le, or blocks, can be
saved in a more ecient format.
The PS Series controller manages the data reduction schedule, which denes when the data reduction process begins and ends, and
how frequently data reduction is performed. The NAS array controller manages the data reduction process.
NOTE: To congure and enable the data reduction feature on a NAS container, you must have group administrator
(grpadmin) privileges.
User quotas are enforced based on logical data size, not on the physical footprint of the data. If two users each have a copy of the
same 1GB le, the physical footprint of the le is 1GB after data reduction, and both users use 1GB of their individual user quota. If
data reduction is disabled, the user quota will continue to be enforced based on the logical space utilization until all the reduced data
is rehydrated.
Data reduction occurs:
During the time and days specied in a schedule.
When a le has not been accessed or modied for the duration of time dened by the le lters in the data reduction policy.
NOTE: Enabling data reduction on a container permanently removes the snapshot reserve functionality for that
container. Snapshot reserves cannot be enabled even if you later disable data reduction. Snapshots can still be manually
created, deleted, and restored, and automatically deleted when the maximum number of snapshots to keep is exceeded.
Data reduction is accomplished through two methods:
Deduplication — Applied whenever data reduction is enabled. Deduplication replaces qualied duplicated data with a pointer to a
single copy of the data.
Compression — Applied only when compression is enabled as part of the data reduction policy. Compression occurs within les
when individual blocks can be saved in a more ecient format than originally written.
Data reduction is performed on blocks in 128KB chunks. Space savings from data reduction are reported in MB units.
Deleting a reduced le will not result in additional free disk space on the container, unless the le being deleted is the last copy of
that le. Free space gained from deleting the last copy of a le will not be available until the disk cleaner process cleans the le
marked for deletion.
When you enable data reduction on a NAS container, default data reduction policy settings are applied to that container. You can
modify the default policy settings or dene other policy settings. The data reduction policy species if data compression is applied
and determines if les qualify for data reduction based on the last time they were accessed and modied. For more information
about setting default values, see Creating Default Data Reduction Properties of a NAS Cluster.
Data reduction aects system performance. When a reduced le is accessed, the controller reads a copy of the deduplicated data
and uncompresses the data as it is being sent to a client. Unless the reduced data is modied, it remains reduced on the disk. After
data reduction is disabled, if at least 5GB of free space is available on the container, when data is read on the container, the data is
uncompressed.
When deciding whether or not to enable data reduction, consider the storage resource and data reduction processing eciency of
your system.
220
NAS Container Operations