Hi there.
I am having trouble understanding different graphical execution plans for one of my SELECT queries. I have tested performance of this query using different joins and indexes and have noticed that the "subtree cost" could increase significantly even though the query execution time decreases with different scenarios. I assumed (incorrectly) that as query time decreases, so would the subtree cost. Why is this not the case? Can anyone give me a rough definition of "subtree cost" and how it relates to execution time?
Thanks for your input.
I am having trouble understanding different graphical execution plans for one of my SELECT queries. I have tested performance of this query using different joins and indexes and have noticed that the "subtree cost" could increase significantly even though the query execution time decreases with different scenarios. I assumed (incorrectly) that as query time decreases, so would the subtree cost. Why is this not the case? Can anyone give me a rough definition of "subtree cost" and how it relates to execution time?
Thanks for your input.
