Since we need to support data sync with every new release of the data sets, we have some issues in tracking veteran records - there is no ID field for veterans or cemeteries, so we are using a combination of fields as an ID. right now those fields are {firstName-lastName-birthDate-deathDate-cemeteryID} as veteran ID, with Cemetery ID: {cemeteryName-address-zip}.

This ensures we populate the database only with complete data. If we don't use birth/death date or cemetery id as keys, we get a lot of duplicate records.
On the other hand, this does mean we will miss a lot of records that have partial data.
Is there anything that can be done about the data source (like having record ids)? @kbowerma