Archives of the TeradataForum
Message Posted: Fri, 17 Aug 2001 @ 14:47:04 GMT
I created 5 duplicate copies of a table with a single column primary index defined as an integer. When I join these tables together on their common primary index, the optimizer treats everything as an ample local process (no redistributions.)
Then, I created 5 duplicate copies of a table with a single column primary index defined as a CHAR. When I join these tables together on their common primary index, the optimizer redistributes the data after every merge join.
Can someone explain this to me?
I have collected stats on all tables. The optimizer didn't care whether the CHAR primary index was case specific or not.
I can send you all of the table ddl, dml, and explains if you would like it.
|Copyright 2016 - All Rights Reserved|
|Last Modified: 28 Jun 2020|