Archives of the TeradataForum
Message Posted: Wed, 18 Oct 2000 @ 09:45:08 GMT
A question on dimensional modelling:
I have a dimensional model in the form of a star schema, the fact table currently holds about 6 million records and has a unique primary index made up of the 8 dimension tables. the dimension tables are relatively small (100's of records) apart from the customer table with about 50,000 records.
I have collected statistics on all the primary indexes (fact table and dimension tables), as well as for each index used in the query. The tables have been designed so that the joining fields are of the same data type and size.
I was hoping that the optimizer would produce a product/merge join plan that would product join all the smaller tables first before merging with the fact table
At the moment the explain suggests that the optimizer will join to the fact table to early on, before having joined to all the smaller tables
any comments /suggestions welcomed
|Copyright 2016 - All Rights Reserved|
|Last Modified: 28 Jun 2020|