@ Eugene: amazing, that technique!

Hi,

I think you should put the first query in a temp table:

SELECT EmpId,

EmpName,

ProcessName,

GroupCode,

PlannedDays,

ActualDays,

DelayedDays,

rowid = ROW_NUMBER() OVER (PARTITION BY EmpID ORDER BY GroupCode)

INTO #base

FROM EmpProcessDetail

Then use this temp table for...

yesterday I did an edition upgrade from SQL 2008 R2 Standard to SQL 2008 R2 Enterprise.

User databases are not affected. Even master database is not affected.

We have our windows files...

using udf's to get column data is a bad idea for performance.

It works well for some rows, but it works loop based, so not using the database engine strength.

So for...

indeed, that worked. Tx

CREATE TABLE [dbo].[printing_header](

[print_batch_number] [bigint] IDENTITY(1,1) NOT NULL,

[creation_dt] [datetime] NULL,

[last_change_dt] [datetime] NULL,

[requester_user_code] [varchar](10) NULL,

[owner_code] [varchar](10) NULL,

[warehouse_code] [varchar](10) NULL,

[language_code] [varchar](10) NULL,

[printing_code] [varchar](10) NULL,

[session_trans_id] [bigint] NULL,

[printing_id] [bigint] NULL,

[confirm_requested] [bit] NOT NULL,

[printing_handler] [varchar](5) NOT...

2009-12-08 00:13:20.38 spid4s Node:1

2009-12-08 00:13:20.38 spid4s KEY: 6:72057595571732480 (3300bbfdce97) CleanCnt:3 Mode:X Flags: 0x0

2009-12-08 00:13:20.38 spid4s ...