Hello everyone, I'm trying to figure out why I'm having a performance issue wth the tOracleInput component.
The job looks like this: tOracleInput --> tMap(with one small lookup) -->tMap(with a really small lookup) --> tOracleOutputBulkExec The tOracleInput read rates is at 34 rows / s and considering that I have about 9 millions rows to read it will take a while... The options I have checked: - tOracleInput Use Cursors for 20 000 rows On the tMap I tried to store the data on the disk without any success. I also launch the job from the TAC with a 10gb min memory so memory is not an issue. On the database I gathered the stats on both schema and in the job I am using a single connection. Anyone have any idea ?
try eliminating all the possilbe issues by setting up a job with just the following: tOracleInput-->tJavaRow leave the tJavaRow empty and run the job to see how fast you can retrieve data. If you're still seeing 34r/s then the issue is most likely in your DB/network layer. If it speeds up a ton, then check the tmap lookup model (having it set to reload each row could cause the issue)