Historical inter-system replication

Historical data is transferred between sites by the Datapump application. Unlike the RealTime replication server, Datapump is only used for replication between systems.

Any Historical data that is inserted or modified at the Hot service on a system is replicated to the other system(s).

Note: There may be one or more destination systems, depending on the configuration of AVEVA Enterprise SCADA. For instance, a system might have an emergency backup system and a decision support system.

The applications that use Historical insert data only at the Hot service. When the data source system changes as a result of dual-host failover, or as a result of a DistribuSyS mode switch, there may still be many outstanding updates. These updates continue to be replicated even though the source system is no longer controlling the data points. If the source system is down, these updates will be replicated upon recovery of the system. Also at this time, data collected at the new source system during the failure will be sent to the recovered system to resynchronize the databases.

Datapump is configured on a per-table basis so that subsets of the tables in a database can be transferred to a destination system. For instance, the Timeline.collect table does not need to be transferred to a decision support system if only hourly timeline data is required.

Note: Datapump is configured using AVEVA tools and predefined configurations.

Under no circumstances should the commands to configure Datapump be used directly. If manual commands are issued to modify Datapump, these commands will be lost if the system is ever reinstalled or recovered after a failure. Issuing manual commands to modify Datapump could also result in inconsistencies and cause replication to fail. To modify Datapump configuration, contact your Service manager.

Datapump maintains a constant connection to both the source and destination databases. It performs small, frequent scans of key tables in the Historical database to extract modifications. A key table is a small table of keys that acts as a queue for data that needs to be transferred. Data transfer through the open connection imposes a small, constant load on the network. A connection failure generates an alarm. For a complete list of Datapump alarms, refer to the AVEVA Enterprise SCADA HMI Controller documentation.