Forum Replies Created

Viewing 15 posts - 16 through 30 (of 39 total)

  • RE: heap or clustered

    GilaMonster (4/12/2009)


    radek (4/12/2009)


    BTW: table like product_type, product,product_group,failure_cause etc.. is small max 2000rows and tbl_FAILURES_archive_only_first_entrys is about 4000, I think thats very small to create indexes but statistics is better...

  • RE: heap or clustered

    GilaMonster (4/12/2009)


    I suspect the cluster vs heap discrepancy is more the result of a lot of bad and missing indexes.

    It would really help if you could post the actual execution...

  • RE: heap or clustered

    RBarryYoung (4/11/2009)


    Wow. Do you have the version of the SQLPLAN that has only 3 logical IOs?

    Yes, It is plan with 3 IO on table tbl_product_area 🙂

    there is result from...

  • RE: heap or clustered

    GilaMonster (4/11/2009)


    Post the exec plan for each please.

    While a heap is occasionally beneficial, in most cases it's better for a table to have a cluster. There's probably a missing index...

  • RE: heap or clustered

    What's your query, what's the table structure in the two cases?

    Table is realy simple:

    CREATE TABLE [dbo].[tbl_PRODUCT](

    [ID] [int] IDENTITY(1,1) NOT NULL,

    [PRODUCT] [varchar](10) NOT NULL

    ) ON [PRIMARY]

    but query is difficult because in...

  • RE: heap or clustered

    RBarryYoung (4/11/2009)


    Yes, just create your table with a Primary Key that is not clustered.

    Right , but how? I tried it but it every create clustered index 🙂

    like this:

    CREATE TABLE...

  • RE: Index tuning

    GilaMonster (4/11/2009)


    Grant Fritchey (4/11/2009)


    Instead, if you were trying to use a covering index to solve the bookmark lookup, all the columns needed to satisfy the query would have to...

  • RE: Index tuning

    Hmm, I don't know that sql 2000 have not include index, that's pity.

  • RE: Index tuning

    MR (4/6/2009)


    The query is this:

    select * from nodes

    where macaddress = '00-13-d3-b4-47-ab'

    I can't save the plan, because its sql server 2000 - but basically it is something like this:

    Select (0%) <-...

  • RE: after rebuilt is fragmentation higher then before

    Ok, thanks I add this condition into procedures.

    BTW: Of course I know, if I have small table that it is not good to create index.. but i dont know that...

  • RE: after rebuilt is fragmentation higher then before

    But let me ask you what made you to defrag your indexes were you experiencing any performance problems or is it a part of your maintenance plan? Since its a...

  • RE: after rebuilt is fragmentation higher then before

    Table is not big, about 20 000 rows with 15 columns, index has 18 pages, and depth=2.

    What cause that it cant fragmentation? only few pages?

    thanks Radek

  • RE: Performance Issues

    Hi I have other question about indexes:-)

    I read something about indexes, that indexes is good create in "big table"

    But when I create table in MS SQL by GUI and define...

  • RE: Performance Issues

    I can asked how did we find column to indexes?

    I know when you use "where" than is good to create index...for big table.

    And I read about profiler,...

  • RE: Performance Issues

    So I have another question.

    For find indexes column is better use DMV? or Profil?, of course I can create indexes where

    I use "word" where or order by..

    I think...

Viewing 15 posts - 16 through 30 (of 39 total)