Like most indexing decisions, it depends on the specifics of your
application. Clustered indexes are most useful for range queries, ORDER
BY and retrieval of a single row based on a unique clustered index key.
Since one doesn't usually select by uniqueidentifier range or ORDER BY a
uniqueidentifier value, a clustered index on a uniqueidentifier value
may not be appropriate since there may be a better choice for your
clustered index. However, a clustered index on a uniqueidentifier
primary key can be beneficial when your primary method of retrieval is
single rows by uniqueidentifier value.
Indexing focuses on SELECT performance but you need to consider DML
performance as well. Due to the random nature of GUIDs under Windows
2000 and above, INSERT performance can be negatively affected with large
tables and a clustered index on a uniqueidentifier value.
Hope this helps.
SQL Server MVP
SQL FAQ links (courtesy Neil Pike):
"Miroo_news" <poczta.fm> wrote in message