Merge Join's row count estimation doesn't consider the case that the join key degenerates to other condition · Issue #57691 · pingcap/tidb · GitHub
More Web Proxy on the site http://driver.im/
You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For example, if we have table t(a int, b int, key a). And the SQL select ... from t t1 join t t2 on t1.a=t2.a and t1.b=t1.b. To use merge join, we'll only regard a as the join key and the t1.b=t2.b will become a normal filter after we finish the join key matching.
But, from the picture, you can see that the row count estimation and the cost calculation don't consider such a case.
The text was updated successfully, but these errors were encountered:
Enhancement
For example, if we have table


t(a int, b int, key a)
. And the SQLselect ... from t t1 join t t2 on t1.a=t2.a and t1.b=t1.b
. To use merge join, we'll only regarda
as the join key and thet1.b=t2.b
will become a normal filter after we finish the join key matching.But, from the picture, you can see that the row count estimation and the cost calculation don't consider such a case.
The text was updated successfully, but these errors were encountered: