A FlashCopy image is an instantaneous copy of a set of data, taken at a particular point in time. FlashCopy can be performed at the volume level, copying data from the source volume to a target volume.
How does FlashCopy work? First, you allocate space in the target volume for
the FlashCopy image.
Then, system applications are paused for only an instant—long enough for
the processor card to make a copy of the internal maps that show where data
blocks are located on the DDMs. After FlashCopy is initialized, and applications
are running again, the storage controller uses the maps to copy the data. Setting
up FlashCopy is greatly simplified if you use the DS Storage Manager intuitive
graphical interface and configuration wizards.
After setting up FlashCopy, you can choose one of several options. The full
— or background — copy makes a physical copy of all the data blocks
associated with the source volume. Administrators generally take this approach
when the target volume is intended for long-term use.
The incremental copy makes a physical copy of only those data blocks that have
been added since the full copy was created.
The copy-on-write approach makes a physical copy of data blocks as they are
updated. This is less taxing on the internal bandwidth of the storage controller
and is generally done when the target volume is intended for temporary use.
Additional FlashCopy functions provide you with even more flexibility, including
multiple-relationship FlashCopy, consistency-group FlashCopy, and data-set FlashCopy
functions (which are only supported for IBM System z and zSeries volumes).