You can add disks or array LUNs to an aggregate so that it can provide more storage to its associated volumes.
This procedure should not be used for aggregates composed of root or data partitions.
Following these best practices when you add storage to an aggregate optimizes aggregate performance:
The new RAID group does not have to be exactly the same size as the existing RAID groups, but it should not be less than one half the size of the existing RAID groups.
Doing so results in the added disks being the target for a disproportionate percentage of new data, causing the new disks to become a performance bottleneck.
If you are adding disks with a different checksum than the aggregate, as when creating a Flash Pool aggregate, or if you are adding disks to a mixed checksum aggregate, you must use the -checksumstyle parameter.
If you are adding disks to a Flash Pool aggregate, you must use the -disktype parameter to specify the disk type.
You can use the -disksize parameter to specify a size of the disks to add. Only disks with approximately the specified size are selected for addition to the aggregate.
If you specify the -raidgroup parameter with a value of new, the storage is added to a new RAID group; this is generally the best way to add storage to an aggregate. You can also specify an existing RAID group with the -raidgroup parameter. If you are adding SSDs to the SSD cache of a Flash Pool aggregate, you do not need to specify the RAID group name; the SSD RAID group is selected by default based on the type of the disks you are adding.
disks specifies the disks to be added in one of the following ways:
If possible, you should use the diskcount option rather than selecting specific disks. Doing so allows Data ONTAP to optimize the disk selection for your configuration.
If you are adding disks to a mirrored aggregate and you are specifying disk names, you must also use the -mirror-disklist parameter.