Just saw this one:

prvreddy (3/12/2009)

---

Hi all,

1)Why a table have maximum one clustered index ?

2)Please give me basic definitions of clustered and non-clustered indexes!

3)What is fragementation and defragmentation?

Please help me...

GilaMonster (3/12/2009)

---

I don't agree there. They...

Do you have a basic idea of what you want to do, and you're looking for suggestions/critique on it, or do you need to hire someone to design the database...

No. ...

I've tuned a lot of databases, sped up a lot of queries for a lot of applications, over the last 9 years. I have yet to find the index...

JacekO, one error. SQL Server can nest transactions, to an extent. That's what @@Trancount is for.

Try this to see what I mean:

if object_id(N'tempdb..#T') is not null

drop table #T;

create...

Doesn't matter. Gus, G, G2, GSquared, even Hey You. It all works. 🙂

At 1-million, the syscolumns cross join took 297 milliseconds CPU, 291 total.

Created a million-row BigNumbers table. 235 CPU, 243 total, using that.

Hemalatha (3/12/2009)

---

GSquared thanks for the reply, but can you be more specific.

You mean to say that

If I have a table A, that has a primary key which is not being...

31 milliseconds CPU and total for that one. Five runs, no real variation (32 ms one time out of five).

Selecting from an actual Numbers table, 0 CPU, 14 total....

Yes. Views don't really store data on their own. They're just a way to connect to the table(s) they query.

Just ran this test:

set statistics time off;

set nocount on;

declare @N int;

set statistics time on;

;WITH

L0 AS (SELECT 1 AS C UNION ALL SELECT 1), ...

Another possible solution:

create table #T (

ID int primary key,

StartDay int,

StartTime char(5),

EndDay int,

EndTime char(5));

insert into #T (ID, StartDay, StartTime, EndDay, EndTime)

select 1,1,'21:10',1,'21:50' union all

select 2,1,'22:48',1,'23:15' union all

select 3,1,'23:50',2,'00:15';

select

datediff(

minute,

cast(starttime as...

There are a number of solutions for something like this. Which one is best depends a lot on what the data means and what it's for.

For example, if it's...

I think I'd be more inclined to use the one that stacks newids together than the one that actually picks more random character strings. Less likely to end up...