It is very much interesting when I get request from blog reader to re-write my previous articles. I have received few request to rewrite my article SQL SERVER – Union vs. Union All – Which is better for performance? wi.th examples. I request you to read my previous article first to understand what is the concept and read this article to understand the same concept with example.
DECLARE @Table1 TABLE (Col INT)
INSERT INTO @Table1
SELECT 1
INSERT INTO @Table1
SELECT 2
INSERT INTO @Table1
SELECT 3
INSERT INTO @Table1
SELECT 4
INSERT INTO @Table1
SELECT 5 /* Create Second Table */
DECLARE @Table2 TABLE (Col INT)
INSERT INTO @Table2
SELECT 1
INSERT INTO @Table2
SELECT 2
INSERT INTO @Table2
SELECT 6
INSERT INTO @Table2
SELECT 7
INSERT INTO @Table2
SELECT 8 /* Result of Union operation */
SELECT Col ‘Union’
FROM @Table1
UNION
SELECT Col
FROM @Table2 /* Result of Union All operation */
SELECT Col ‘UnionAll’
FROM @Table1
UNION ALL
SELECT Col
FROM @Table2
GO
The difference between Union and Union all is that Union all will not eliminate duplicate rows, instead it just pulls all rows from all tables fitting your query specifics and combines them into a table.
A UNION statement effectively does a SELECT DISTINCT on the results set. If you know that all the records returned are unique from your union, use UNION ALL instead, it gives faster results.
If you look at the resultset it is clear that UNION ALL gives result unsorted but in UNION result are sorted. Let us see the query plan to see what really happens when this operation are done.
From the plan it is very clear that in UNION clause there is an additional operation of DISTINCT SORT takes place where as in case of UNION ALL there is no such operation but simple concatenation happens. From our understanding of UNION and UNION ALL this makes sense.
There are three rules of UNION one should remember.
- A UNION must be composed of two or more SELECT statements, each separated by the keyword UNION.
- Each query in a UNION must contain the same columns, expressions, or aggregate functions, and they must be listed in the same order.
- Column datatypes must be compatible: They need not be the same exact same type, but they must be of a type that SQL Server can implicitly convert.
Reference : Pinal Dave (https://blog.sqlauthority.com)
19 Comments. Leave new
Hello,
I wish I had read this article one day before.
Yesterday when I was working on one production issue, I saw a strange case, in which we create two temporary table on the fly, and then we do a select statement with a union. Every time we use to get an error… You wont believe we spent 2-3 hours on it. At last we saw the temporary tables that were created with select * into command was creating different datatype, one with int and another with char, and union was not able to perform its action and our process was failing. I learnt a good lesson, it was worth it :)
Thanks,
Imran.
Also beware of implicit convertions
select 10
union all
select getdate()
Its a hobby to read your blog’s article on daily basis to refresh my knowledge.
Some tricky part which mostly ignore, Catch it from your articles.
Keep up the good work.
God Bless you
Joggee.
First time i am going through this site.It’s simply the best.
U have all the details without redundant details. Thanks tons for this.
I want to know the difference between UNION and INSERT INTO
is ther a way to use a union all statement and then issue a where at the end to filter the entire query?
example:
table 1
tcol1 tcol2 tcol3
1 1 1
2 2 1
table 2
tcol1 tcol2 tcol3
3 3 1
4 4 2
SELECT tcol1,tcol2 FROM (
SELECT tcol1,tcol2
UNION ALL
SELECT tcol1,tcol2
) AS temptable
WHERE tcol3=’1′ <—
thanks
sorry for the dumb question, i should have included tcol3 in my inner query…
Is it possible to use ‘union all’ in hibernate, Since I am using Java and Hibernate for my application, I need to convert the query using ‘union all’ to hibernate.Please Help me to solve this issue.
Hi Ria,
Did you ever get a solution for this? What alternative did you use?
I was looking at performance gaining and a friend of mine doing a peer review of my code suggested that we use union all instead of union, I was not sure how I will gain performance and while researching came across your article.
I do need a distinct list of records and therefore used union
but she has an opinion that I do a union all and then group by the columns individually before applying union all.
Which is best?
get the distinct list of table 1 (by using group by clause) and distinct list of table2 and then do union all
or select list from table 1 and select list from table 2 and do union there by eliminating duplicates in individual tables
@Aparna,
Why don’t you look at Execution plan. Look at cost of each step and each query. You will get an answer.
Also, please post your observation.
~ IM.
I was short of time with lots of work in hand so did not really look at the execution plan. I will post my findings in this forum. Thanks Imran for your suggestion. wondering if anyone had similar issues.
FYI Table 1 is not distint by itself and Table 2 is not distinct by itself but there are dateid’s in Table1 and Table2 that makes their union distinct
TABLE1
DATEID INVESTOR
20090530 1
20090530 2
20090530 1
TABLE 2
DATEID INVESTOR
20090630 1
20090630 2
20090630 1
Any one has thoughts until I look at the execution plan
Hi
I have a question .
I need to use select statement and use two from clauses.
that is, everything is same in both the select statements and from clause except there is an extra join in one from clause. what i am trying to do is use condition and take the reqired from clause but i could not do it.
Any help plz
@navi
please show the query so someone can help you with it.
how to give descriptive headers for each column we pull from the database for the fixed length query.
Thanks for the post. . It helped a lot. keep the good work up..
thanks
These can not be used to combine sequence values. Refer this post for more information
id
1
2
3
4
5
6
can I get this output without using union???