Posted by Erland Sommarskog on 04/17/06 11:13

Alexander Kuznetsov (AK_TIREDOFSPAM@hotmail.COM) writes:
> Serge Rielau wrote:
>> Struggling to research bookmarks I ran into:
>> http://msdn2.microsoft.com/en-US/library/ms180920(SQL.90).aspx
>> "Bookmark Lookup is not used in SQL Server 2005. Instead, Clustered
>> Index Seek and RID Lookup provide bookmark lookup functionality."
>> Cool, let's bypass that one.

I think this is mainly a terminology issue. Instead of talking of
bookmark lookups, they present it as a join between the NC index and
the clustered index. But there is no real difference between SQL 2000
and SQL 2005, as far as I understand.

> Interesting. My understanding is that SQL Server 2000 does not store
> clustering factor as part of statistics, and the reason is simple: it
> makes no sence for bookmark lookups. However, for more efficient RID
> lookups, it would make perfect sence to both calculate clustering
> factor as part of statistics gathering and have the optimizer use it.
> Yet I tried to google up "clustering factor"+"SQL server" and that
> brought up nothing relevant to SQL Server.
>
> Does SQL Server 2005 optimizer use clustering factor, and if yes,
> what's the proper word for it?

I'm not really sure what you mean with clustering factor. As for RID
lookups, they only occur with heaps, that is tables without clustered
indexes.


--
Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se

Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/prodtechnol/sql/2005/downloads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinfo/previousversions/books.mspx

[Back to original message]