Question

I have a SQL stored procedure in which one statement is taking 95% of the total time (10 minutes) to complete. #Records has approximately 133,000 rows and Records has approximately 12,000 rows.

  -- Check Category 1 first
  UPDATE #Records
    SET Id = (SELECT TOP 1 Id FROM Records WHERE Cat1=#Records.Cat1)
    WHERE Cat1 IS NOT NULL

I have tried adding a index to Cat1 in #Records, but the statement time did not improve.

CREATE CLUSTERED INDEX IDX_C_Records_Cat1 ON #Records(Cat1)

A similar statement that follows, takes only a fraction of the time

  -- Check Category 2
  UPDATE #Records
    SET Id = (SELECT TOP 1 Id FROM Records WHERE Cat2=#Records.Cat2)
    WHERE ID IS NULL

Any ideas on why this is happening or what I can do to make this statement more time effective?

Thanks in advance.

I am running this on Microsoft SQL Server 2005.

Answer 1

update with join maybe

update t
set t.ID = r.ID
FROM (Select Min(ID) as ID,Cat1 From Records group by cat1) r 
INNER JOIN #Records t ON r.Cat1 = t.cat1 
Where t.cat1 is not null