help with an sql problem

• ChrisM@Work

  SSC Guru

  Points: 186127

  More actions

  August 17, 2011 at 9:39 am

  #1369694

  Sure.

  SELECT

  d.[rec_index],

  d.file_desc,

  d.is_modified

  FROM (

  SELECT

  [rec_index],

  file_desc,

  is_modified,

  rn = ROW_NUMBER() OVER(PARTITION BY file_desc ORDER BY rec_index DESC)

  FROM MyTable

  ) d WHERE d.rn = 1 AND d.is_modified = 0

  ^{“Write the query the simplest way. If through testing it becomes clear that the performance is inadequate, consider alternative query forms.” - Gail Shaw}

  For fast, accurate and documented assistance in answering your questions, please read this article.
  Understanding and using APPLY, (I) and (II) Paul White
  Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden

• ChrisM@Work

  SSC Guru

  Points: 186127

  More actions

  August 17, 2011 at 9:57 am

  #1369710

  Heh! Funny - I thought your solution had a triangular join but it isn't. That >= ALL () appears to be very efficient with this small data set. It's twice as fast as my solution and the nearest logical alternative to yours:

  select *

  from (

  select *

  from #MyTable as t2

  where rec_index = (

  select MAX(rec_index)

  from #MyTable as t1

  where t2.file_desc = t1.file_desc)

  ) as t3

  where t3.is_modified = 0

  Definitely worth scaling this up to a proper test.

  ^{“Write the query the simplest way. If through testing it becomes clear that the performance is inadequate, consider alternative query forms.” - Gail Shaw}

  For fast, accurate and documented assistance in answering your questions, please read this article.
  Understanding and using APPLY, (I) and (II) Paul White
  Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden