This section provides the following information:
Data replication enables controlled migration of production services from a primary cluster to a secondary cluster either in the event of a disaster or as part of a planned procedure. Data is continuously replicated from the primary cluster to the secondary cluster either synchronously or asynchronously, or a combination of both, depending on the recover point objectives of the application services that are supported by the clusters.
The Geographic Edition framework supports the following software for data replication:
Availability Suite feature of Oracle Solaris
EMC Symmetrix Remote Data Facility software
Hitachi TrueCopy and Universal Replicator software
Oracle Data Guard software, where the Oracle database is running on a local Oracle Solaris Cluster node or on a remote system that is not running Oracle Solaris Cluster software
Oracle GoldenGate software
Oracle Solaris ZFS snapshots feature
Oracle ZFS Storage Appliance
Geographic Edition script-based plug-ins
Availability Suite and Oracle Solaris ZFS snapshots are host-based data replication facilities which replicates data at the file system or logical volume level within the operating system. Hitachi TrueCopy and Universal Replicator software, EMC Symmetrix Remote Data Facility software, and Oracle ZFS Storage Appliance software use a storage-based data replication facility which replicates data at the storage system level and provides a transparent service to applications. Oracle Data Guard software, Oracle GoldenGate software, and MySQL software are application-based data replication facilities that maintain one or more standby databases as synchronized replicas of a production database.
The Availability Suite feature of Oracle Solaris software is a host-based replication facility that replicates disk volumes between geographically separated primary clusters and secondary clusters in real time. Remote mirror replication enables data from the master volume of the primary cluster to be replicated to the master volume of the geographically separated secondary cluster through a TCP/IP connection. A remote mirror bitmap tracks differences between the master volume on the primary disk and the master volume on the secondary disk.
The remote mirror software continually replicates the data to remote sites while your applications are accessing the data volumes. You can also manually update the data on the secondary site volume by issuing a command to synchronize the primary and secondary site volumes. You can also restore data from the secondary volume to the primary volume by issuing a command to reverse-synchronize the volumes. For more information about Availability Suite, see the Availability Suite man pages such as dscfg(1M).
EMC Symmetrix Remote Data Facility software provides remote storage replication for disaster recovery and ensures data and systems availability through remote site failover. EMC Symmetrix Remote Data Facility devices are configured in pairs. The mirroring relationship between the pairs becomes operational as soon as the EMC Symmetrix Remote Data Facility links are online. The EMC Symmetrix Remote Data Facility global memory stores information about the pair state of operating EMC Symmetrix Remote Data Facility devices. For information about EMC Symmetrix Remote Data Facility software, see the product documentation.
Hitachi TrueCopy and Universal Replicator software is a storage-based replication facility that provides a host-independent data replication for geographically separated clusters. Hitachi TrueCopy and Universal Replicator software enables the primary volumes to remain online for all hosts and while processing both read and write I/O operations. If a disaster or system failure occurs, the secondary copy of the data can be run for a recovery with minimal loss of data. For more information about Hitachi TrueCopy and Universal Replicator software, see the product documentation.
MySQL database software has a built-in replication component that is an application-based replication facility. MySQL replication keeps one or more standby databases in sync with the active master. Geographic Edition supports the use of MySQL replication between two databases. Only asynchronous replication is available.
You add a plug-in for MySQL replication to a protection group. This plug-in monitors the status of the replication and controls the replication direction. For more information about MySQL software, see the product documentation.
Oracle Data Guard software is an application-based replication facility that maintains one or more standby databases as synchronized replicas of a production database. Geographic Edition supports the use of Oracle Data Guard replication where the Oracle database is running locally on an Oracle Solaris Cluster node or remotely on a separate system that is not running Oracle Solaris Cluster software. When the Oracle database runs locally, the HA for Oracle Database or Support for Oracle RAC data service must be used. When the database runs remotely, the HA for Oracle External Proxy data service must be configured on the local cluster for the remote database.
You add an Oracle Data Guard broker configuration that is controlled by the Oracle Data Guard software to a protection group. The Geographic Edition framework creates a shadow server proxy resource group for each Oracle Data Guard broker configuration. The shadow server proxy resource group “shadows” the database resource group to manage and monitor the Oracle databases that are under the control of Geographic Edition framework. For more information about Oracle Data Guard software, see the product documentation.
Oracle GoldenGate is a data replication software for databases. It takes changes from a source database and replicates them to a target database. Oracle GoldenGate consists of the following processes:
The manager controls the other processes, including the start and stop functions.
The extract process pulls data out of a database and stores it either locally or remotely. If the extract process has to store the data locally, it does so by writing it to a file. If the extract process has to store the data remotely, it communicates with a remote server process and gets the server to store the data in files. The files that contain the data are called trails.
The server process is started automatically by the manager process, if a remote process contacts the manager to send data. Once the server process is started, communication takes place only between the sending extract process and the server process. The manager ceases to have any role in the communication between the sending extract process and the receiving server process. The server process takes the data from the sending extract process and stores it into a remote trail. Currently, there is no interface in the Oracle GoldenGate software to stop a server process. Hence, these processes must be killed manually.
A replicat process picks up the information from a trail and stores it into a target database. To achieve this, the replicat process must be able to make a local connection to the database.
A healthy Oracle GoldenGate instance has at least the manager up and running. Although such an instance is valid, it cannot process any data. A fully working Oracle GoldenGate instance has at least an extract process or a replicat process with the corresponding server up and running. You need at least two Oracle GoldenGate instances; one to serve as a sender and the other to serve as the receiver.
You can have different kinds of source and target databases. For example, you can have replication between MySQL and Oracle Database in either direction. The replication itself is always asynchronous. Synchronous replication is not possible using Oracle GoldenGate. The Oracle GoldenGate replication can be configured as either bi-directional or uni-directional.
For more information about Oracle GoldenGate, see the product documentation.
Oracle Solaris ZFS snapshots is a host-based replication facility that enables ZFS dataset snapshots to be transmitted from the primary cluster to the secondary cluster. Oracle Solaris ZFS has this built-in feature to take consistent snapshots of a ZFS dataset, which can be transmitted and replayed on another system that hosts a copy of the ZFS dataset. This feature of ZFS provides a method of replicating ZFS datasets within a disaster recovery (DR) environment and enables the primary cluster to transmit ZFS snapshots of a ZFS dataset to a secondary cluster. For more information about Oracle Solaris ZFS snapshots, see Chapter 8, Working With Oracle Solaris ZFS Snapshots and Clones in Managing ZFS File Systems in Oracle Solaris 11.3.
Oracle ZFS Storage 7000 appliances support snapshot-based replication of projects and shares from a source appliance to any number of target appliances manually, on a schedule or continuously. The replication includes both data and metadata. Remote replication, or just replication, is a general-purpose feature optimized for disaster recovery, data distribution, disk-to-disk backup and data migration. For more information about Oracle ZFS Storage 7000 appliances, see the product documentation.
The Geographic Edition framework provides a generic interface module referred to as script-based plug-ins. With this module, you can configure custom scripts for additional replication technologies to use in a Geographic Edition configuration. For more information about Geographic Edition script-based plug-ins, see Chapter 13, Script-Based Plug-Ins in Oracle Solaris Cluster 4.3 Geographic Edition System Administration Guide.
This section provides the following information:
Resource groups and data replication components enable the Geographic Edition framework to manage data replication and takeover between clusters. For more information about configuring data replication, see the guide for the data replication product you are using:
The Geographic Edition framework extends Geographic Edition resource management features to integrate the data replication products. When you configure a protection group, the Geographic Edition framework creates replication resource groups for monitoring and controlling data replication.
A data replication component is the replication object that Geographic Edition framework monitors and manages. The data replication component is added to a Geographic Edition protection group, which can then start, stop, switch over, or take over the replication component. The Geographic Edition framework coordinates these actions between the clusters and the applications being protected.
Types of data replication components include Availability Suite device groups, Oracle Data Guard broker configurations, Oracle ZFS Storage Appliance replicated projects, and EMC Symmetrix Remote Data Facility device groups. For information about configuring data replication components in Geographic Edition configurations, see the guide for the data replication product you are using.