Recently I had an application developer approach me and asked if I could provided him with a list of sprocs by returned row count. We had an issue where the application passed in a number of parameters which attempted to return 200k plus rows of data and the application was timing out. He changed the required parameters in the application and a reasonable number of rows were returned as expected. Short term solution to this one problem.
However there are always timeout issues with this particular application and we got to thinking that maybe other sprocs that were called using parameters would also fail at some point in time because too much data was being returned.
His request made sense if we could proactively identify potential issues with sprocs returning large number of rows.
I vaguely remember doing something like this before, but couldn't find my code snippet so I was wondering if anyone else had done this before and had some idea where to point me.
I tried stats and cached plans but the information wasn't in these dm views as far as I could tell.
However there are always timeout issues with this particular application and we got to thinking that maybe other sprocs that were called using parameters would also fail at some point in time because too much data was being returned.
His request made sense if we could proactively identify potential issues with sprocs returning large number of rows.
I vaguely remember doing something like this before, but couldn't find my code snippet so I was wondering if anyone else had done this before and had some idea where to point me.
I tried stats and cached plans but the information wasn't in these dm views as far as I could tell.