raw trace file interpretation

[satish]

Dear Erman,

Recently,i came up with a performance issue with a request which is running for the first time.We have enabled the trace and got the trace file.

while the request was running,i have been manually checking the wait events on which the session was waiting and found for almost 4 hours,session keep on switching between cache buffer chains and latch:shared pool.The total runtime for request is about 10 hours

But when i used the below command to check the elapsed time for these wait events,i am getting a very less value almost less than 10minutes.

cat PROD1_ora_24787.trc | grep WAIT | grep 'cache buffer chains ' | awk '{sum += $6}END{print sum}'

Our development team later changed the code and the runtime was reduced,but i would like to know below information.

From your experience,can you please share some method with which we can quickly identify the total elapsed time for a specific wait event from raw trace file?

Thank you for all the support

[ErmanArslansOracleBlog]

why dont you use tkprof?
The easiest way is to look at the very last line in the TKPROF output, which will mention the total time captured in the trace: look for the phrase "elapsed seconds in trace file"