I am joining two large tables, 100+million rows each. They each have a clustered index on 1 column and I am joining those 2 columns together. I was thinking this join should be instant, but its taking longer than I expected.
The query looks like this:
FROM Fact.TableName a
INNER JOIN Journal.TableName j ON j.ID = a.Journal_ID
The execution plan is doing a index scan on each table (not on the clustered index) and then doing a hash match on the results.
In reality I want a bunch of columns back from "a" and a handful from "b". What would be the ideal index design in this example?Craig
View Complete Post