Inner join vs separate statements

Question

I am trying to get some more context on this out of curiosity. So far when I run 2 separate sql statements I find in SQL Profiler that I have no CPU cycles, less reads and less duration than taking the script and using Inner join. Is this a typical case, I am looking for help to understand this better.

Simple example:

SELECT *  FROM dbo.ChargeCode 

SELECT *  FROM dbo.ChargeCodeGroup

vs

SELECT *  
FROM dbo.ChargeCode c 
INNER JOIN dbo.ChargeCodeGroup cc ON c.ChargeCodeGroupID = cc.ChargeCodeGroupID

From what I guess is that inner join cost extra CPU cycles because its doing a nested loop. Am I on the right track with this?

Answer 1

The simple answer is that you're doing two different things here. In your 1st example you're retrieving 2 separate entities. In your second example, your asking the RDBMS to combine (join) 2 entities into a single result set.

A join is one of the most powerful capabilities of an RDBMS - and it will (usually) do it as efficiently as it possibly can - but that's not to say it's free or cheap.

Answer 2

SELECT * FROM sometable

must scan whole table. If there are indexes on ChargeCodeGroupID column on either table, it will be much faster for INNER JOIN to only scan index. (By their name, I guess there are). Of course, if there is no index on either ChargeCodeGroupID column, second query is slower than the first one.