ok no issue i try this ...

before i explain something in my process ..

BCP happens in both table (A1,A2) .. A1 ( key column ) has identity ......

if i replicate the table i have to do modify the whole application .. that is hard ... in one sp i have one select qry . this sp executed...

if i create index after BCP , further i do deletion and updation .. so it will reduce performance due to cluster index creation .

barry solution :

CREATE [UNIQUE] INDEX idxTablenameKeyDT

ON TableName ( Key, DT )

with only column KEY and selected a column in multiple place .. for that and...

IS IT POSSIBLE TO ANALYZING ABOUT RAID , FILEGROUP IN DATABASE ... I AM NOT AWARE OF THAT ... SO ONLY ASKING ???

INDEX :

=============

NC_INDEX1_A1 > nonclustered located on PRIMARY > KEY, QNBR, TIMESTAMP, BUS_CNTR

Table definition is difficult .. that has 1030 columns ... also...

this result is without hint .. hint shown 25 % sort cost .. so i dropped ...

wud like to solve without hint ..

Table 'A1'. Scan count 8, logical reads 160572, physical reads 0, read-ahead reads 128208.

Table 'A2'. Scan count 8, logical reads 305858, physical reads 3, read-ahead reads 320673.

i am not aware...

ss the sort cost newly add to 25 %..

to run statistics on .. without giving the qry output can we take the statistics output ??.. like execution plan can...

if i use the option ( merge join ) .. one of the table goes index scan .. another one is table but cost is reduced from 24...

i cannot use index on DT column .. if i try to create index , have to create in lot places ... in one place with this same join select...

DT is not a index column . can i force index ??? if so what is the syntax ??

this both table has 1020 columns ...

there is no relation ship .. constraint .. only this 2 column is having non cluster index .. nothing else ..

and...

How to give permission 'view system state' in sql 2005 ???

Thank you ...

can u say how to find the current compatibility level in database???

will it set for each database or server level ?