brief description:
Referring to this article:https://community.qlik.com/t5/Support-Knowledge-Base/Handling-Duplicate-Rows-when-doing-a-Full-Load-... and the new option introduced in version NOV2021SR1 of qlik replicate materialized by ‘ Prevent creation of duplicate records on the target’ in the task settings , an improvement could be added to the select statement executed behind this option 'prevent ..'.
value proposition for the idea:
Instead of the actual statement that is issued on the source database (DB2 Z/OS ) to filter rows on primary key cols ,an extra clause ‘optimize for X rows could be added to this sql statement. This will improve the dynamic access path build on the source database for retrieving rows on the specific table and will avoid qlik replicate to make an tablespace scan on data rows which sometimes ends in a abnormal way for huge table
Target audience (user role) for the idea:
users that has db2 Z/OS as source in data replication
How does this idea help the user?
This idea will help the users by really get rid of hidden passthrough filters which are still necessary in cases you have big tables to load from db2 Z/OS even when activating option in the load 'Prevent creation of duplicate records on the target'
Additional context to help in the review process:The statement behind option Prevent creation of duplicate records on the target could integrate the statement optimize for ..like written in this passthrough filter example:
- 1=1 ) ORDER BY POITEC,POIVER FOR READ ONLY OPTIMIZE FOR 1000000 ROWS
This will improve access path on the source database and avoid troubles with db2 workfiles when loading huge table.
thanks