SQL 2000 does not have a perfect query optimizer.

My query is against a tera byte  plus of data and the joins in question contain millions of rows. I have...

thanks

or

select *

from table1

join (table2 join table3

on tbale2.col = table1.col

trying to improve performance of query

several multi million row tables joing together want to experiment with join order structure.

i thought you could pre join certain chunks of the from clause

somethig...

for whats it worth franks function runs fastest in my sql express inastallatin on my creaking lap top

(11000 row(s) affected)

(100000 row(s) affected)

--===== Frank's function

DBCC execution completed. If DBCC printed...

if your using sql 2005 which i suspect you might be given your name! you should use try catch blocks which are much neater and much efficient as you dont...

you database design isnt right. You shouldnt be storing the same data in two different tables becasue you are likely to have problems with consistency.

why not just add a bit...

'so prioritise those in a certain order.' Not sure what you mean by this. This code will run fast. Much much faster than a cursor.

Version above only distinguishes between transaction on different days. Version below will do it down to milli seconds which is probably what you want.

---DDL--

if

---DDL--

if

object_id('test_donations') is not null