Oracle IO testing
When I first saw the suffix BATCHED in an Oracle execution plan from 12c, I was curious to see what is hidden behind and how it works. I had some spare time for testing and wanted to share my findings with you here.
This posts finalizes a few weeks of work on comparing SLOB and ORION IO testing tools.
In this blog post, I will share the physical IO testing results I got by running SLOB and ORION on the same system and on the same disks. I will use those results as a reference in a few blog posts to come. As of now, I would like to make few points based in the results here.
As SLOB takes the IO testing on Oracle Instance level, you need to watch for several things before you declare official testing results. My suggestion is: Don’t take any result for granted. You need to assess those before stating that you are done.
If you are wondering what I have been busying myself with, this post explains it. As you may have noticed, I am still testing one of the Oracle systems using the SLOB framework and learning on my way. I ran several tests with the same parameters (Readers 24) and noticed that for one reason or another awr.txt reports different runtimes.
I was wondering why I was getting inconsistent runtimes for similar conditions. The answer to the mystery was very simple: The runit.sh script just generates awr.txt reports for the last and before last awr snapshots.