GSquared (8/6/2012)

---

And the key to the article was the bit about inline 1X1 UDFs as "scalar" functions with better performance, and the point about Set Stats. Regardless of ICaps...

Jeff Moden (8/6/2012)

---

Heh... make no "Butts" about it. 🙂 I know, I know... you've had...

Jeff Moden (8/5/2012)

---

mmilodragovich (8/1/2012)

---

Thanks for the detailed and METHODICAL analysis. I once was troubleshooting a performance problem in a large, complicated application and it came down to a "set...

Can I safely assume that you have...

@peter-757102 Sorry for the misunderstanding.

Just re-read the article and realize that I have missed some points. Must be sleepy while reading it the first time :crazy:

GSquared (8/2/2012)

---

Usman Butt (8/2/2012)

---

@GSquared Any reason for not turning it into iTVF/iSF? I guess then the it could...

peter-757102 (8/2/2012)

---

I did some timings on my own using the code in the article and as expected the time the baseline and the iTVF take are identical. Not surprisingly I...

Using GSquared's testing technique (tweaked the insert script a bit to avoid optimizer's table spool intelligent duplicate string handling)

--DATA POPULATION SCRIPT

CREATE TABLE ##T

(

...

Sorry, since not everyone have the tally table physically on their servers, here is a version using cteTally

SET ANSI_NULLS ON

GO

SET QUOTED_IDENTIFIER ON

GO

CREATE FUNCTION [dbo].[InitialCapiTVF_CteTally] (@String VARCHAR(8000))

RETURNS TABLE WITH SCHEMABINDING AS...

GSquared (8/1/2012)

---

Here's a function I tested, and the test I used:

USE [ProofOfConcept]

GO

/****** Object: UserDefinedFunction [dbo].[InitialCapTest1] Script Date: 08/01/2012 11:20:42 ******/

SET ANSI_NULLS ON

GO

SET QUOTED_IDENTIFIER ON

GO

ALTER FUNCTION [dbo].[InitialCapTest1]...

As always great article. But I am sure there must be something up your sleeve 😉 After following you so much I am not believing that you would be using...

That policy is certainly possible, but why you need that is somewhat not understandable. There are several factors to be considered before implementing the compression on the table.Anyhow, the sys.partitions...

Jeff Moden (5/22/2012)

---

I'm pretty sure that I stated that the code in the article is NOT the correct solution for VARCHAR(MAX) because just changing the input datatype to VARCHAR(MAX) causes...

michael vessey (5/23/2012)

---

have you validated that there is no open source code in the CLR that has a reciprocal license that requires you to publish the source code of your...

SQL Kiwi (5/22/2012)

---

David Benoit (5/22/2012)

---

I think the issue here though Paul is that you just stuck a language into the database realm where the people that support it had no...