Question

SELECT A,B,C,D from Table1 Where Constraint1

vs

SELECT A from Table1 Where Constraint1
SELECT B from Table1 Where Constraint1
SELECT C from Table1 Where Constraint1
SELECT D from Table1 Where Constraint1

Is there any performance difference ?

Answer 1

I think that the first one is faster, but.. let’s test it!

Considering in the second group of selects to return A, B, C & D.

IN SQL SERVER 2008 R2:

Let’s try it:

Create Table Test 
(
A int not null,
B int not null,
C int not null,
D int not null,
)

insert into Test values (1,2,3,5)

Running Execution plan for the first:

select A, B, C, D from Test

and this results on:

It’s on spanish, but you can see the numbers. Has a CPU Cost of 0.0000796

So, let’s try it with the second:

select A from Test
select B from Test
select C from Test
select D from Test

and this results on:

As you can see, each select use the same CPU Cost, but.. here has to do 4 selects. With a little row, you won’t see the difference, but, let me update, with more data

UPDATE

Now, I made a couple of inserts, and now we have:

select count(*) from test

76183 rows

Let’s do the same: We got the first select, and in the execution plan, we got:

as you can see, we got more CPU COST (of course, more rows!)

And then let’s check the 4 selects:

Again, we got the same CPU COST, but we run it 4 times !!

In conclusion, there is no difference on showing one field or four, but we got a big difference on doing one select against 4.

I know it’s a simply question, but I wanted to do some “science/research”