Replication Agent for SAP HANA creates shadow tables and triggers at the primary database; the triggers move changes from user table to shadow tables, then a replication scanning process moves the changed data from shadow tables to the replicate database.
Replication Agent for SAP HANA offers several replication strategies.
When you use this strategy, tables are scanned separately without any commit order. Only primary key information and operation type is stored in the shadow tables; the data in the other database is queried from the primary table when it is applied to the replicate database.
Use this strategy when performance and load at the database server is more important than immediate data consistency.
This strategy is a variant of the eventual consistency strategy. The only difference is that staging tables, rather than user tables, are created and populated in the replicate database. Replication Agent for SAP HANA also manages the removal of data from the CDC tables. This strategy shares eventual consistency's performance and load characteristics.
Use this strategy when you want to stage the data before writing it to the final destination, and when replication performance and primary database server load considerations are more important than immediate data consistency.
When you use this strategy, updates are applied at the replicate database in the same order as they were applied at the primary database. To achieve consistency, transaction identifiers are recorded with each database operation, and all columns are captured with every change.
Use this strategy when data consistency is more important than performance and database server load. Data consistency between the primary and replicate database server varies according to the amount of latency in the replication path.
This strategy places the heaviest load on the primary database and does not perform as well as the eventual consistency strategy.
This strategy is a variant of the transactional consistency strategy. The only difference is that staging tables, rather than user tables, are created and populated in the replicate database. Replication Agent for SAP HANA also manages the data removal from the CDC tables. This strategy shares transactional consistency's performance and load characteristics.
Use this strategy when you want to stage the data before writing it to the final destination. You can also use it when immediate data consistency is more important than performance and primary database server load.