Hi, Scenario overview ------------------ We are testing an application prototype and inserting 1000 rows at a time from 1 table and populating into 2nd table in a loop and inserting 1M rows. 2nd table have primary key on 2 column, 1 column unique key and 2 non-unique indexes each having 6 columns. This is 10g on WIN2K. I created the tablespaces with uniform extent size and allocate sufficient minextents to table and all indexes so that no need to allocate more extent. Now when I start inserting for first 2 runs transaction per secons(tps) inserted are around 25000-30000 but suddenly on 3rd run I see tps drop to 18000 and then sometimes 30000 and other times 18000. I didn't see any new allocation for tables and indexes, no log switches since the file is large enough, no increase in temp tablepspace increase, some waits but that are for both when tps was 30000 and 18000. What can be the possible reason for such difference in tps numbers with running same script? Wait events looks like: . 98% of db file sequential read wait is on file containing indexes for table. Elapsed times include waiting on following events: Event waited on Times Max. Wait Total Waited ---------------------------------------- Waited ---------- ------------ db file sequential read 5041 0.50 59.86 log file sync 208 1.01 1.18 log buffer space 136 0.00 0.14 Thanks --Harvinder ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx put 'unsubscribe' in the subject line. -- Archives are at //www.freelists.org/archives/oracle-l/ FAQ is at //www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------