Is there a way to check the load a Qlik replicate task puts on source database system? We currently have a Production and QA environment each running a similar task sourcing from the same Oracle database (production instance). Would like to try and understand the load these tasks would have on the source database. The source is an Oracle database running on linux. We need to shut down the QA tasks so as to have only one task running against the source.
The tasks for both QA and PROD have target endpoints of logstream with separate tasks to write to their associated targets.
We should be able to create addition QA endpoints in our PROD environment and create tasks writing the same PROD data to QA.
Please consult your Oracle DBA to inquire about the load on the source database, as they possess the statistics prior to and following the task execution.
During Full Load, the load on the source database will be substantial, depending on the number of tables you intend to perform a Full load on simultaneously. Conversely, for Change Processing, the load will be reduced since Replicate will access your Redo logs and retrieve the changes.
Replicate is going to read the Logstream Audit files and there will be an increment in your IO reads on the Replicate server. Also, the 7 new tasks you created will occupy CPU and Memory. It depends on your server capacity and I would say there won't be much performance impact.
Please consult your Oracle DBA to inquire about the load on the source database, as they possess the statistics prior to and following the task execution.
During Full Load, the load on the source database will be substantial, depending on the number of tables you intend to perform a Full load on simultaneously. Conversely, for Change Processing, the load will be reduced since Replicate will access your Redo logs and retrieve the changes.
Will check with the DBAs. Regarding log stream, our production environment currently has a task reading from the source database to log stream. There are currently 7 child tasks that then read from log stream and write to our production Azure environment. Would there be any performance decreases if we were to duplicate those child tasks and target them to our QA Azure environment (so a total of 14 tasks reading from log stream)?
If it is only for CDC then there won't be any performance impact on your source DB as you are reading the changes from the Logstream staging path. There won't be a performance decrease even if you have 14 child tasks.
But if you try to do a Full Load then there will be a performance impact on your source database.
My question was on the Qlik side, not source database. We currently have 7 Qlik tasks reading from the log stream and targeting our production Azure environment. If we were to increase this to 14 tasks reading from log stream (current 7 tasks plus 7 new tasks that would write data to our QA Azure environment), would there be a performance impact on Qlik replicate?
Replicate is going to read the Logstream Audit files and there will be an increment in your IO reads on the Replicate server. Also, the 7 new tasks you created will occupy CPU and Memory. It depends on your server capacity and I would say there won't be much performance impact.
