User journal replication of IFS objects, data areas, data queues - assure_mimix - 10.0

Assure MIMIX Administrator Reference

Product type
Product family
Assure MIMIX™ Software
Product name
Assure MIMIX
Assure MIMIX Administrator Reference
First publish date

MIMIX supports the ability to replicate IFS objects, data areas, and data queues   either by user journal (database) replication or by system journal (object) replication processes. User journal replication of these objects is sometimes referred to as advanced journaling and is a form of cooperative processing.

Replicating IFS objects, data areas, and data queues through the user journal can be more efficient than replication processes based on the system journal. Each time a journaled IFS object, data area, or data queue changes, only the changed bytes are recorded in the journal entry.

Default values for new installations and new data group object entries automatically configure data group object entries that identify *DTAARA and *DTAQ objects for user journal replication. Default values for IFS objects configure these objects to use system journal replication, however you can specify data group IFS entries to allow user journal replication. When object or IFS entries allow user journal replication, MIMIX uses tracking entries to uniquely identify each configured object.

When a data group has optionally enabled a journal-centric configuration for *DTAARA or *DTAQ object types, or for IFS objects, all of those objects that are journaled to the user journal associated with the data group are replicated by user journal replication processes unless they are excluded by data group entries. You do not need to configure data group object entries or IFS entries because the journaled objects are tracked internally to MIMIX, but MIMIX will create tracking entries for each journaled object that is internally tracked.

The same data group can be used to replicate cooperatively processed files, data areas, data queues, and IFS objects, as well as to replicate other object types and IFS objects that are not cooperatively processed. However, there may be practical reasons to have separate data groups for some types of data. You may need to consider how much data is replicated through the same database apply session.

When journaled IFS objects, data areas, and data queues are replicated from the user journal, they are always assigned to database apply session A. When files are cooperatively processed, the apply sessions used depend on whether the data group is configured for single-threaded or multithreaded database apply processing. When multithreading is configured, files are assigned to apply session B. When single-threading is configured, files can be assigned to any apply session, including session A. In a data group that uses single-threading, you may need to consider whether any transactions need to be serialized with database files. For more information, see Planning for journaled IFS objects, data areas, and data queues.