IBM® InfoSphere® Replication
Server distributes,
consolidates, and synchronizes data for high availability, high throughput,
and business continuity.
Two types of replication, Q replication and SQL replication, support
a broad range of business scenarios:
- Q replication
- A high-volume, low-latency replication solution that uses WebSphere® MQ message queues
to transmit transactions between source and target databases or subsystems.
A capture process reads the DB2® recovery
log for changes to source tables and sends transactions as messages
over queues, where they are read and applied to targets. Q replication
offers the following advantages:
- Minimum latency
- Changes are sent as soon as they are committed at the source and
read from the log.
- High-volume throughput
- The capture process can keep up with rapid changes at the source,
and the multithreaded apply process can keep up with the speed of
the communication channel.
- Minimum network traffic
- Messages are sent by using a compact format, and data-sending
options enable you to transmit the minimum amount of data.
- Asynchronous
- The use of message queues enables the apply process to receive
transactions without needing to connect to the source database or
subsystem. If either of the replication programs is stopped, messages
remain on queues to be processed whenever the program is ready. Because
the messages are persistent, the source and target remain synchronized
even if a system or device fails.
Q replication supports DB2 for z/OS® and DB2 for Linux, UNIX, and Windows as source platforms. Q replication
supports the following target platforms: DB2 for z/OS, DB2 for Linux, UNIX,
and Windows, Informix®, Microsoft SQL Server, Oracle, and Sybase.
- SQL replication
- SQL replication captures changes to source tables and views and
uses staging tables to store committed transactional data. The changes
are then read from the staging tables and replicated to corresponding
target tables. SQL replication offers the following advantages:
- Capture once
- With staging tables, data can be captured and staged once for
delivery to multiple targets, in different formats, and at different
delivery intervals. Both tables and views are supported as sources.
- Flexibility
- You can replicate continuously, at intervals, or for one time
only. You can also trigger replication with database events. You can
replicate a subset of the table by excluding columns and filtering
rows. You can also use expressions and other functions to transform
data before it is applied.
- Hub-and-spoke configurations
- You can replicate data between a master data source and one or
more replicas of the source. Changes that are made to the master source
are propagated to the replicas, and changes that are made to the replicas
are also propagated to the master source. Whenever a conflict occurs
between the data that is sent from the master source and data this
is sent from a replica, the data from the master source takes precedence.
SQL replication supports the following source and
target platforms: DB2 for z/OS, DB2 for Linux, UNIX, and Windows, IBM DB2 for IBM i, Informix, Microsoft SQL Server, Oracle, and Sybase.
In addition, SQL replication supports Teradata targets.