Comments on: Adaptive Cursor Sharing with SQL Plan Baselines – Bind Sensitiveness

By: Sql Plan Mangement(SPM) and Adaptive Cursor Sharing(ACS) : My résumé | Mohamed Houri’s Oracle Notes

Mon, 01 Apr 2013 09:41:02 +0000

[...] article written by one of those modest and smart Oracle guys Coskan Gundogar which he has entitled Adaptive Cursor Sharing with SQL Plan Baselines – Bind Sensitiveness. Finally, I have ended up my “SPM-ACS collaboration Giro” with the Optimizer blog article [...]

By: anthonydba

anthonydba — Wed, 02 May 2012 11:07:11 +0000

I do agree with Dom Brooks as I have seen personally many such a performance issues due to bind variable peeking and ACS can’t address it at the very first execution of the SQL with a poor plan.I have a history of SQLs used to run in seconds took many hours due to different bind values with poor plan.
To me,ACS is helpless during those fire fighting time.

By: When dbms_xplan.display_sql_plan_baseline fails to show the plan « Coskan’s Approach to Oracle

Wed, 11 Apr 2012 20:45:22 +0000

[...] is how he did it with a little bit of polishing. (I will be using the baseline generated on this post – dbms_xplan.display_sql_plan_baseline works for this but not important for the post [...]

By: Adaptive Cursor Sharing with SQL Plan Baselines « OraStory

Adaptive Cursor Sharing with SQL Plan Baselines « OraStory — Sat, 24 Mar 2012 08:13:18 +0000

[...] And the answer to that is no. Which I’m glad Coskan has shown in his follow-up post. [...]

By: Dom Brooks

Dom Brooks — Fri, 23 Mar 2012 12:10:43 +0000

They do work fine together.

But I know when I read the documentation, I came away with some questions that I wanted to look at. I wondered if they did more together than they do.

For example, did baselines preserve some of the ACS not-so-secret sauce such that if the ACS information aged or was flushed out of the cache, we didn’t have to repeat the multiple executions to get the bind awareness back?

And the answer to that is no. Which I’m glad you’ve shown above.

1. The bind awareness will kick back in.
2. The baseline mechanism will allow the optimizer to use the ACS feedback and consider both plans in the baseline.
3. But we need to repeat the executions to get back i.e. three runs to have one bind aware plan, four runs to have the two bind aware plans that we preserved initially.

I know it’s not necessarily a massive deal is it?
We’ve had to do one extra execution each of the statement but if that initial execution with the “wrong” plan was a big enough performance problem, then the implications could be significant.

And the two mechanisms that I thought might be designed to work together to address it, don’t.