I beg to differ. A comma is NOT an ANSI join, CROSS JOIN would be!

You could also see if your tables could benefit from table partitioning.

stephane3d (6/18/2008)

---

Hum... I think I miss a thing here.

If I have an indexed view, should I get rid of any indexes on the table? Is the performance cost on the...

Joe Celko (6/18/2008)

---

You used singular table names but they don't seem to have only one row. You used reserved words for column names or vague useless names like "organization"...

If it has "pre"-aggregated data, you bet!!!

Try:

SELECT [Year], [Month] ,[Org1],[Org2],[Org3]

FROM

(SELECT U.Organization, R.Users , [Year], [Month]

FROM #Report R INNER JOIN #USERS U ON U.UserAlias = R.UserAlias) AS src

PIVOT ( SUM (Users)...

cast(a.im as varbinary(max)) = cast(b.im as varbinary(max))

Is not that bad, right ?

declare @jb_name nvarchar(128)

declare c cursor fast_forward for

select name from msdb.dbo.sysjobs

open c

fetch next from c into @jb_name

while @@fetch_status = 0

begin

print 'changing : ' + @jb_name...

Can you post DDL for...

In addition to Gila's post you should check for missing indexes that may help reduce the number of scans.

Did you try the "indexed view" idea ?

What version of the ODBC Driver are you using ?

FROM BOL:

... BINARY_CHECKSUM ignores columns of noncomparable data types in its computation. Noncomparable data types include text, ntext, image, cursor, xml, and noncomparable common language runtime...

Or simply CAST to varbinary(max) and compare

declare @a table( id int, im image )

declare @b-2 table ( id int, im image )

insert into @a (id, im) values (...

Hopefully your developers are not using select *... If that is not the case you can't do much 😉