How you use SSD storage pools

To enable SSDs to be shared by multiple Flash Pool aggregates, you place them in a storage pool. After you add an SSD to a storage pool, you can no longer manage it as a stand-alone entity—you must use the storage pool to assign or allocate the storage provided by the SSD.

You create storage pools for a specific HA pair. Then, you add allocation units from that storage pool to one or more Flash Pool aggregates owned by the same HA pair. Just as disks must be owned by the same node that owns an aggregate before they can be allocated to it, storage pools can provide storage only to Flash Pool aggregates owned by one of the nodes that owns the storage pool.

If you need to increase the amount of Flash Pool cache on your system, you can add more SSDs to a storage pool, up to the maximum RAID group size for the RAID type of the Flash Pool caches using the storage pool. When you add an SSD to an existing storage pool, you increase the size of the storage pool's allocation units, including any allocation units that are already allocated to a Flash Pool aggregate.

You can use only one spare SSD for a storage pool, so that if an SSD in that storage pool becomes unavailable, Data ONTAP can use the spare SSD to reconstruct the partitions of the malfunctioning SSD. You do not need to reserve any allocation units as spare capacity; Data ONTAP can use only a full, unpartitioned SSD as a spare for SSDs in a storage pool.

After you add an SSD to a storage pool, you cannot remove it, just as you cannot remove disks from an aggregate. If you want to use the SSDs in a storage pool as discrete drives again, you must destroy all Flash Pool aggregates to which the storage pool's allocation units have been allocated, and then destroy the storage pool.