<i>How would we know which index is been used during Native IO reads...</i> You look at the programming to see what index was specified. Native I/O does exactly what you tell it to do. If the programming is for index A, then that's the index that will be used. <i>...or SQL query execution...</i> For SQL, it's a bit harder to tell. It can change on different days depending on index statistics, past usage patterns, the DB2 version/release and possibly other factors. Run the process in debug to see what the resulting messages tell you. Usually, the same indexes will be used; it just isn't guaranteed. And, of course, appropriate indexes can often be determined simply by looking at the SQL and at the available indexes to see if any are an obvious match. Tom

<i>How would we know which index is been used during Native IO reads...</i> You look at the programming to see what index was specified. Native I/O does exactly what you tell it to do. If the programming is for index A, then that's the index that will be used. <i>...or SQL query execution...</i> For SQL, it's a bit harder to tell. It can change on different days depending on index statistics, past usage patterns, the DB2 version/release and possibly other factors. Run the process in debug to see what the resulting messages tell you. Usually, the same indexes will be used; it just isn't guaranteed. And, of course, appropriate indexes can often be determined simply by looking at the SQL and at the available indexes to see if any are an obvious match. Tom