Chapter 2 • Synchronizing at a Glance

Incremental Synchronizations

You run an incremental synchronization on a link to determine which records in the endpoints were created or modified since the previous synchronization task, and then synchronize data between the two endpoints according to the mapping and settings you defined for the link.

➤In an incremental synchronization, the Synchronizer first sends a request to each endpoint to supply a list of records created or modified since the time indicated by the time stamp for the link. The time stamp for a link is the time the last synchronization task started running. The endpoint supplies this list by comparing the time stamp for the link with the time each record was last modified as recorded by the endpoint. If a record was modified after the time indicated by the time stamp for the link, it is included in the list.

➤If a record that is included in the list for one of the endpoints is not previously known to the Synchronizer, and therefore does not appear in the identity map table for the link, the Synchronizer infers that the record was created after the last synchronization. If you enabled record creation in the other endpoint, the Synchronizer creates a corresponding record in that endpoint and updates the identity map table to include this correspondence.

➤If a record in the list already appears in the identity map table for the link, and the record’s version is different from the version in the identity mapping record, the Synchronizer infers that the record has been updated since the last synchronization. The Synchronizer then performs a cycle redundancy check to determine if fields included in the mapping were modified and require synchronization. If synchronization is necessary, the Synchronizer updates the corresponding record in the other endpoint, depending on the settings for the link.

41