Veritas FlashSnap Point-In-Time Copy Solutions 5.0 AdministratorÆs Guide, HP-UX 11i v3, First Edition, May 2008

11Point-in-time copy solutions
Software used in point-in-time copy scenarios
Persistent FastResync can track the association between volumes and their
snapshot volumes after they are moved into different disk groups. After the disk
groups are rejoined, persistent FastResync allows the snapshot plexes to be
quickly resynchronized.
For more information, see the Veritas Volume Manager Administrator’s Guide.
Instant volume snapshots
The traditional type of volume snapshot that was provided in VxVM is of the
third-mirror type. This name comes from its original implementation by adding
an additional plex to a mirrored volume. After the contents of the third-mirror
(or snapshot plex) had been synchronized from the original plexes of the
volume, it could be detached as a snapshot volume for use in backup or decision
support applications. Enhancements to the snapshot model allowed snapshot
volumes to contain more than a single plex, reattachment of a subset of a
snapshot volume’s plexes, and persistence of FastResync across system reboots
or cluster restarts.
Release 4.0 of VxVM introduced instant snapshots, which offer advantages over
traditional third-mirror snapshots. The benefits of instant snapshots include
immediate availability for use, quick refreshment, and easier configuration and
administration. Full-sized instant snapshots are similar to third-mirror
snapshots in that they are the same length as the original volume.
Space-optimized instant snapshots require less space than full-sized snapshots
by recording changed regions in the original volume to a storage cache. As the
original volume is written to, VxVM preserves its data in the cache before the
write is committed.
For more information, see the Veritas Volume Manager Administrator’s Guide.
Disk group split/join
One or more volumes, such as snapshot volumes, can be split off into a separate
disk group and deported. They are then ready for importing on another host
that is dedicated to off-host processing. This host need not be a member of a
cluster but it must have access to the disks on which the volumes are configured.
At a later stage, the disk group can be deported, re-imported, and joined with the
original disk group, or with a different disk group.