Update copy operation, Read and write i/o operations – HP StorageWorks XP Remote Web Console Software User Manual
Page 35

If the journal-obtain operation starts at the primary data volume, the primary storage system obtains all
data of the primary data volume as the base-journal data, in sequence. The base-journal contains a
replica of the entire data volume or a replica of updates to the data volume. The base-journal will be
copied from the primary storage system to the secondary storage system after the secondary storage
system issues a read-journal command. After a base-journal is copied to the secondary storage
system, the base-journal will be stored in a restore journal volume in a restore journal group where the
secondary data volume belongs. After that, the data in the restore journal volume will be restored to
the secondary data volume, so that the data in the secondary data volume synchronizes with the data
in the primary data volume.
The base-journal data is stored in the entire data volume or the area for the difference. The area for
the difference is used when the difference resynchronization operation is performed. The journal data
for the entire data volume is created when the data volume pair is created. The difference journal
data is obtained when the pair status of the data volume changes from the Suspending Pair resync
status to the status. Merging the difference bitmaps that are recorded on both primary and secondary
data volumes enables you to obtain the journal data for only difference. When a data volume pair is
suspended, the status of data that is updated from the host to the primary and secondary data volumes
is recorded to the difference bitmap.
The base-journal data of primary storage system is stored to the secondary storage system journal volume
according to the read command from the secondary storage system. After that, the base-journal data is
restored from the journal volume to the secondary data volume. The initial copy operation will finish
when all base-journals are restored.
If you manipulate volumes (not journal groups) to create or re-synchronize two or more data volume pairs
within the same journal group, the base journal of one of the pairs will be stored in the restore journal
volume, and then the base journal of another pair will be stored in the restore journal volume. Therefore,
the operation for restoring the latter base journal will be delayed.
You can specify None as the copy mode for initial copy operations. If the None mode is selected,
initial copy operations will not be performed. The None mode must be used at your responsibility
only when you are sure that data in the primary data volume is completely the same as data in the
secondary data volumes.
Update Copy Operation
When a host performs a write I/O operation to a primary data volume of a data volume pair, an update
copy operation will be performed. During an update copy operation, the update data that is written to
the primary data volume is obtained as an update journal. The update journal will be copied to the
secondary storage system, and then restored to the secondary data volume.
The primary storage system obtains update data that the host writes to the primary data volume as
update journals. Update journals will be stored in journal volumes in the journal group that the primary
data volume belongs to. When the secondary storage system issues "read journal" commands, update
journals will be copied from the primary storage system to the secondary storage system asynchronously
with completion of write I/Os by the host. Update journals that are copied to the secondary storage
system will be stored in journal volumes in the journal group that the secondary data volume belongs
to. The secondary storage system will restore the update journals to the secondary data volumes in the
order write I/Os are made, so that the secondary data volumes will be updated just like the primary
data volumes are updated.
Read and Write I/O Operations
When a primary storage system receives a read I/O for a URz primary data volume, the primary storage
system performs the read from the primary data volume. If the read fails, the redundancy provided by
RAID-1 or RAID-5 technology recovers the failure. The primary storage system does not read the URz
secondary data volume for recovery.
When a primary storage system receives a write I/O for the primary data volume with PAIR status, the
primary storage system performs the update copy operation, as well as writing to the primary data volume.
The primary storage system completes the primary data volume write operations independently of the
update copy operations at the secondary data volume. The secondary storage system updates the data
Hitachi Universal Replicator for z/OS user guide for XP12000/XP10000 Disk Arrays and SVS 200
35