Logic of data warehouse zipper table

• Full zipper

1、 Extract the full amount of data from the source database , The current day partition stored in the data warehouse source layer table （SA surface ）

2、 Put the data warehouse source layer table into the full volume data of the current day partition , Dump into the current table of the history layer （SH surface ）

3、 Handle the zipper table of the historical layer of the warehouse （SH_H surface ）： It will exist in the zipper table of the historical layer of the warehouse （SH_H surface ） in , But it does not exist in the current table of the history layer （SH surface ） The data of , Perform chain closure （ The opening date remains unchanged , The closing date is the same day ）

4、 Handle the zipper table of the historical layer of the warehouse （SH_H surface ）： It will exist in the zipper table of the historical layer of the warehouse （SH_H surface ） in , And exists in the current table of the history layer （SH surface ） The data of （ This part of data is not updated or newly added , The opening date remains unchanged , The closing date is permanently valid ）,UNION ALL On , Existing in the current table of the history layer （SH surface ） The data of , But it doesn't exist in the history layer zipper table （SH_H surface ） The data of （ This part of data is updated or newly added data , The opening date is the same day , The closing date is permanently valid ）, Open the chain

• Incremental zipper
• Running water