Not from me - I tend to agree with you.

Roy Ernest (7/26/2011)

---

I was wondering what you would advice the value for max degree of parallesim on the server level on an OLTP server with 24 logical CPUs and 256...

I can only speak for myself...

* groan *

Chazz88 (7/26/2011)

---

...By that I mean the rows have been moved around, like what should be row 2 is row 13 and row 4000 is now row 7000 and so on....

can you create soft-NUMA nodes with CPUs 1 and 2?

>> Yes.

If the question had asked:

can you create a soft-NUMA node with CPUs 1 and 2?

>> No.

🙁

The BIT data type can hold the values 0, 1, or NULL.

See http://msdn.microsoft.com/en-us/library/ms177603.aspx

Converting to bit promotes any nonzero value to 1.

See http://msdn.microsoft.com/en-us/library/ms191530.aspx

In many cases, it is possible to write a...

zapouk (7/24/2011)

---

I haven't looked into yet but I was thinking about changing my 'upserts' to using the new merge statement. Is this generally a good idea and if yes, will...

The ISNULL, COALESCE and CHECKSUM methods are unsafe because they can produce incorrect results.

There is a much better way than using ISNULL, COALESCE, COUNT, or CHECKSUM.

http://sqlblog.com/blogs/paul_white/archive/2011/06/22/undocumented-query-plans-equality-comparisons.aspx

Nope, cos that wouldn't be well-formed...

If the tags are well-formed, this will work:

CREATE TABLE MySampleData (ID INT,HTMLText TEXT)

INSERT INTO MySampleData(ID,HTMLText)

SELECT 1,'NoHTML Inside' UNION ALL

SELECT 2, '<Tag1><tag1.1>yada</tag1.1></Tag1>' UNION ALL

SELECT 3 ,'<OBX.5><OBX.5.1>Blah Blah Data</OBX.5.1></OBX.5>Other Stuff outside of...

The Data Loading Performance Guide - it's a in-line link in my previous post.

I'd be interested to see the query plan for the DELETE, because without knowing what the clustered index is on that table, I have no way to know whether it...

We normally use OUTER APPLY, yes. It's not a nice situation, but sometimes (often) necessary.

Generally speaking, the optimizer is...