Details
-
Task
-
Resolution: Not a Bug
-
Major
-
None
-
1.7.0.Final
-
None
-
False
-
False
Description
Hello,
First of all I want to thank all maintainers and contributors for your excellent work on the connector.
We are currently testing it in an UAT Oracle database, and we encounter some performance issues.
DBA said that Debezium Logminer's query was executing for 7 mins / hours approximately and represented 20 % of the DB total load (with 25 Gi of redo logs generated / day - knowing that in production we are talking about 150 Gi of redo logs).
What tools do we have at our disposal to improve perf ?
We are currently testing :
- setting the poll.interval.ms to 1min (it's not clear if the query will always run less often or only when new records exists)
- improving the query by using 1 regex for all tables
Do you have any ideas on how we could improve the situation ?