Home of the ColdFusion Community
|
Mailing Lists
|
Home /
Groups /
SQL
to cluster or to set free
Author: Michael Dinowitz
Short Link: http://www.houseoffusion.com/groups/sql/thread.cfm/threadid:920#3421
By default, the primary key index of any table created in MS SQL is
set to unique and clustered. I see no reason for the PK to be
clustered and usually rebuild the index, removing the cluster from it.
Sometimes I assign it to a field that actually does cluster. Other
times I just leave it alone.
Is there some secret to the use of unique clustered indexes that I'm missing?
Thanks
Michael
Author: Carl Von Stetten
Short Link: http://www.houseoffusion.com/groups/sql/thread.cfm/threadid:920#3422
One important question to ask yourself: Is this table related to other
tables? If so, is the primary key used to join with other tables? If
so, you probably want to leave the primary key index clustered. The
clustered index actually stores all of the table content, so a join
using a clustered primary key index will be faster than joining to a
non-clustered primary key index (because this will require additional
SQL operations to retrieve the rest of the table's content).
HTH,
Carl
On 11/14/2011 3:38 PM, Michael Dinowitz wrote:
----- Excess quoted text cut - see Original Post for more -----
|
May 24, 2013
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
