Hello, so as long as the index is a unique index or primary key, then a full scan of the target tables is avoided for insert and delete operations.
1) What is your definition of high latency?
2) Are your source and target servers' clocks synchronised?
3) The inference is that your target is an Oracle database. Please specify if this is not the case
The target database is often the bottleneck because you may have fewer connections in CDC updating the target database than there are on the source updating the source database.
Have you reviewed the performaning tuning document at
https://www.ibm.com/docs/en/idr/11.4.0?topic=replication-performance-monitoring-tuning ?
------------------------------
Robert Philo
------------------------------
Original Message:
Sent: Mon March 07, 2022 11:53 PM
From: data icons
Subject: CDC high latency
Index is there. Seems that target is applying slower than the source.
------------------------------
data icons
Original Message:
Sent: Thu March 03, 2022 09:38 AM
From: Will Hathcock
Subject: CDC high latency
When I've seen this it's been an index issue where an index was missing.
------------------------------
Will Hathcock
Original Message:
Sent: Wed March 02, 2022 08:35 AM
From: data icons
Subject: CDC high latency
Hi guys, anyone faced high latency on Oracle replication? I've tried increasing global batch size and placed the heavy tables in their own individual subscription. Table structure are the same between source and target. Any clues are much appreciated.
------------------------------
data icons
------------------------------
#DataIntegration
#DataReplication