Forum Replies Created

Viewing 15 posts - 1,306 through 1,320 (of 7,191 total)

  • RE: Deducing from Wait Stats

    I think I'd be happy to go with the recommendations therein.  Don't be afraid to tweak the settings if necessary to improve performance, though.

    John

    August 16, 2017 at 5:01 am

    #1955435

  • RE: sys.master_files and sys.database_files difference

    muzikfreakster - Tuesday, August 15, 2017 9:38 PM

    for one of my critical database , physical name for sys.database_files and sys.master_files doesn't match...

    August 16, 2017 at 4:37 am

    #1955427

  • RE: Deducing from Wait Stats

    Do you have a link to the article, please?  Our standard is as follows:
    ·          For servers that have NUMA configured, MAXDOP should not exceed the number of CPUs...

    August 15, 2017 at 4:40 am

    #1955264

  • RE: Polling and determine the status of a job?

    Nicole

    How about a trigger on sysjobhistory?  I think that's legal, although I've never tried it.

    John

    August 14, 2017 at 9:35 am

    #1955180

  • RE: Deducing from Wait Stats

    CXPACKET just means you have parallelism.  It doesn't on its own mean you have a problem with parallelism.  Please don't blindly reduce parallelism just for the sake of it.  How...

    August 14, 2017 at 9:17 am

    #1955177

  • RE: 'Actual Rows' missing in query plans...

    Arsh - Monday, August 14, 2017 2:13 AM

    GilaMonster - Sunday, August 13, 2017 6:00 AM

    August 14, 2017 at 2:20 am

    #1955113

  • RE: DBCC FREEPROCCACHE

    It may, or it may not.  But if the query covers data that isn't uniformly distributed (lots more Johns than Jims), you're likely to see parameter sniffing.  Setting optimize for...

    August 11, 2017 at 9:16 am

    #1955005

  • RE: 'Actual Rows' missing in query plans...

    No.  If you're getting the plans by querying the DMVs for the plan cache, those don't have the actuals in them at all.  They never do.  They're compiled when the...

    August 11, 2017 at 9:07 am

    #1955003

  • RE: DBCC FREEPROCCACHE

    The easiest conclusion to draw is that you had a poor plan for the query you were running.  Once you cleared the cache, the query recompiled with a better plan...

    August 11, 2017 at 8:42 am

    #1954995

  • RE: Time increments

    Phil Parkin - Friday, August 11, 2017 8:21 AM

    John, Thom, these are not 'calculated fields'. Or perhaps the OP used the wrong...

    August 11, 2017 at 8:27 am

    #1954990

  • RE: Time increments

    WITH Numbered AS (
        SELECT
             FieldSort1
        ,    FieldSort2
        ,    FieldSort3
        ,    ROW_NUMBER() OVER (PARTITION BY FieldSort1 ORDER BY FieldSort2, FieldSort3) AS RowNo
        FROM #TimeIncrements
        )
    SELECT
     ...

    August 11, 2017 at 7:55 am

    #1954977

  • RE: 'Actual Rows' missing in query plans...

    Arshad

    The plans that are stored in cache are plans that are used for all executions of a particular query until the plan is evicted from cache.  That's why...

    August 11, 2017 at 7:46 am

    #1954972

  • RE: Add prefix to results

    t2 36824 - Friday, August 11, 2017 6:25 AM

    When I run this:
    UPDATE Params16
    SET Keycode = LEFT(Keycode,PATINDEX('%!!%', Keycode) + 1)...

    August 11, 2017 at 7:07 am

    #1954961

  • RE: Combinations question with a twist

    You've posted in the SQL Server 2008 forum.  Is that actually what you're using?

    John

    August 11, 2017 at 4:29 am

    #1954939

  • RE: Criteria for creation Histogram Step

    Type "algorithm that SQL server uses to determine the RANGE_HI_KEY" into your favourite search engine.  That's what I did, and the second result looked as if it might contain something...

    August 10, 2017 at 10:38 am

    #1954844

Viewing 15 posts - 1,306 through 1,320 (of 7,191 total)