Monitoring performance of PL/SQL code
Do either 9i or 10g have any out-of-box mechanisms for monitoring the aggregate performance of blocks of code?
For example: let's say I want a report of every table trigger in a schema, the total number of times it has executed in the last 24 hours, the average execution time, min and max execution time, and resource consumption (cpu, disk, etc.) per execution.
Are there any mechanisms to get me close to this?
I know I can turn on SQL tracing, tkprof the results and maybe get some of the data I'm looking for from the tkprof output, but sql tracing seems to focus on SQL alone, with no facility for rolling statistics up to the trigger/procedure/function/etc level.
I think I also can use dbms_application_info to label chunks of PL/SQL code by module, and get some of this data from v$sql or v$sqlarea... although that would require changes to every single trigger, wouldn't it?
Are there any other approaches I've missed?
