Hello Everyone,

I have a requirement where I have a global temp table created with on commit preserve rows option. Now, this table will hold reporting data and the application which uses it might be triggered by hundreds of users simultaneously.Basically the application inserts some data from multiple source tables using joins and after into the temp table and after applyig multiple validations for each session depending upon the user input, the report is generated using the value in the temp table at that time.
I want to know the actual storage of the temp tables- Consider the case when each user simultaneously inserts 10k rows into the temp table and we have total 100 users. Though each user will see only the data relevant to his/her session, is it true that ultimately there would be an overhead of selecting 10k rows from a total of 1 million(10000 * 100) at any point of time for any user?

is it true that ultimately there would be an overhead of selecting 10k rows from a total of 1 million(10000 * 100) at any point of time for any user?

Or is it something like the for each session, the temp table would be created on the fly and would always hold the data specific to that session only and not the cumulative data for all sessions.