Forum Replies Created

Viewing 15 posts - 361 through 375 (of 508 total)

  • RE: Invalid Column Name Error .

    Normally when I see a problem with invalid column name in a #Temp table, there will be a table with the same name already created in the same session. ...

  • RE: Joining two tables need all records from both periods regardless which one has more.

    Your problem is that you are using the transactional data as the left side of the query (the anchor).

    What you need is to have a constant set of data on...

  • RE: Disk I/O

    Before going down the path of insisting that your IO subsystem is inadequate you should be very, very sure that the cause of excessive IO is not poorly optimized queries....

  • RE: SQL Help

    This will give you what you want. The OUTER APPLY will out perform the row number solution if there are many calls per customer and there is an index...

  • RE: SUB-SELECT (I think?)

    iklektic,

    Your post is very similar to a post from a few weeks ago that I used in an article that I submitted and has been approved for publication here on...

  • RE: Performance tunning SQL 2005

    The posted replies are very good. I especially use Gail Shaw's articles and I've found them very useful.

    But even before I go digging into performance issues that I haven't...

  • RE: Group By for Indexed Views

    You won't be able to index a view on this, but I can tell you that using an APPLY with a top 1 can be very fast. Here's an...

  • RE: slow select on MSSQL 2005 cluster symantec

    You probably also want to check for fragmentation on your clustered index. SQL Server will first try to store the LOB (your ntext column) in the data pages. ...

  • RE: FROM (one of several tables, depending on . . . )

    Since you are doing this in a stored procedure anyhow, you can load the data into a temp table first. The stored procedure makes the decision on which table...

  • RE: Self joined table list

    If you have them defined as Foreign Key constraints then this query should point you in the right direction:

    SELECT object_id, OBJECT_NAME(object_id)

    , parent_object_id, OBJECT_NAME(parent_object_id)

    , referenced_object_id, OBJECT_NAME(referenced_object_id)

    FROM sys.foreign_keys

    WHERE parent_object_id = referenced_object_id

    Todd Fifield

  • RE: Mentor

    I completely agree with Jeff on this one. Divide and conquer techniques should be part of any T-SQL curriculum. I do it all the time in stored procedures...

  • RE: Zip Code Radius search - working but have question

    Matt,

    This was quite a while ago and I don't have the actual code I used at hand. Basically you divide the radius (in miles) you want by 111 (carried...

  • RE: Zip Code Radius search - working but have question

    I had to do something like this before SQL 2005 so CTE wasn't available. It was in a stored procedure where I'd get a zip code and number of miles...

  • RE: Discussion: Removing the option to ' SELECT * FROM OBJECT'

    I work a lot with an application where virtually every column in every table is a user defined data type. In stored procedures using temp tables, I can't use...

  • RE: Scalar UDF Performance Issue

    Jeff,

    I'm awaiting that article. It would be nice to have something that looks pretty to resolve these arguments I get into.

    Todd Fifield

Viewing 15 posts - 361 through 375 (of 508 total)