Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
Author |
Topic |
egemen_ates
Yak Posting Veteran
76 Posts |
Posted - 2012-06-12 : 03:12:13
|
my query show belown ; iexamine query plan clusteredIndexScan(viewclustered),how can create nonclustered index for nonclustered index seek.i use which columns for nonclustered index?SELECT TOP 1 (A+B) FROM view_deneme WITH (NOLOCK,noexpand) WHERE A_DATE <= {ts '2012-05-31 00:00:00'} AND M_ID = 14 ORDER BY A_DATE DESC, Y_DATE DESC, K_ID DESC |
|
nigelrivett
Master Smack Fu Yak Hacker
3385 Posts |
Posted - 2012-06-12 : 04:22:59
|
The date will have to scan - although it can us e an index.You might get a seek on M_ID but how selective is it?The order by means that it has to select all the rows and sort them to get the first one so I don't think a seek would be helpful anyway.==========================================Cursors are useful if you don't know sql.SSIS can be used in a similar way.Beer is not cold and it isn't fizzy. |
 |
|
SwePeso
Patron Saint of Lost Yaks
30421 Posts |
Posted - 2012-06-12 : 04:47:13
|
Drop the ODBC style for dates.WHERE A_DATE < '20120531 00:00:00' N 56°04'39.26"E 12°55'05.63" |
 |
|
jackv
Master Smack Fu Yak Hacker
2179 Posts |
Posted - 2012-06-12 : 11:12:50
|
Could you also post your DDL with the existing index definitions with some test data?Jack Vamvas--------------------http://www.sqlserver-dba.com |
 |
|
|
|
|