Forum Replies Created

Viewing 15 posts - 46 through 60 (of 149 total)

  • RE: Remove partitioning

    It sounds like you've described a non-clustered index that is partitioned separate from the base table.  Can you post the definition of the index?

    If that is the case...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: Running date update

    Phil Parkin - Tuesday, February 14, 2017 7:13 AM

    While the SELECT version of this query is appealingly succinct, the UPDATE version is...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: Limiting number of rows to be displayed per page

    Newbi - Friday, December 23, 2016 11:09 PM

    Hi all,Can I have some insight on how to limit the number of rows(x)to be...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: Long running insert

    Johan Fredin - Thursday, January 26, 2017 7:02 AM

    Johan Fredin - Thursday, December 15, 2016 3:51 AM

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: Query tuning - Help needed

    ffarouqi - Thursday, December 15, 2016 1:31 PM

    Thanks Chris. However, the output results from your query doesn't match the output from the query...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: multiple small indexes or large covering

    A very skinny index can read data faster, because more records fit per page.  However, every insert/update/delete will need to be propagated to all of the indexes containing the affected...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: Any situations where a ROLLBACK won't rollback changes against a linked server

    Andrew P - Wednesday, January 11, 2017 4:13 PM

    however, I may have canceled query execution

    Do you still have the transaction open?
    [code...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: t-sql 2012 compare datetime fields

    wendy elizabeth - Thursday, January 12, 2017 9:13 AM

    date1 (datetime,null) to date2(datetime,null).

    WHERE date1 = date2
    OR date1 IS NULL AND date2 IS NULL

    You...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: What data should be converted/formatted in DB, vs in-app?

    There is a distinct difference between converting data and formatting it.

    Formatting should nearly always be done at the presentation layer (application or report).  This generally allows your presentation...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: How to select from 5 lines but only display 3

    jonathanm 4432 - Thursday, January 12, 2017 12:15 PM

    whenriksen - Thursday, January 12, 2017 12:12 PM

    Post...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: Initial size for TempDb data and Log file?

    On a side note, the original Op mentioned his TempDb was stored on his SAN.  IMO, that is a waste of SAN and Backup resources.  We put our tempdb on...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: How to select from 5 lines but only display 3

    Post some DDL and sample data.  Your question looks fairly involved and I don't want to waste my time writing a query against data that doesn't accurately represent your problem.

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: Numbers are double

    I would start by updating the join syntax.  Your duplicate values are almost certainly caused by an incomplete join criteria.  I think the newer syntax makes that easier to notice.

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: Where to place WITH (NOLOCK) ?

    @BOR15K

    To answer your base question, use the NoLock hint in front of each table that you want it applied to. In my opinion, you add maintenance risk to your...

    Wes
    (A solid design is always preferable to a creative workaround)

  • RE: Nonclustered Columnstore with only one rowgroup even with > 1 million rows

    SQL Server will try to pack a rowgroup as full as possible, to the max of 1,048,576. The deltastore will not be converted to a rowgroup until there are...

    Wes
    (A solid design is always preferable to a creative workaround)

Viewing 15 posts - 46 through 60 (of 149 total)