
Signature
Hilary Cotter
Director of Text Mining and Database Strategy
RelevantNOISE.Com - Dedicated to mining blogs for business intelligence.
This posting is my own and doesn't necessarily represent RelevantNoise's
positions, strategies or opinions.
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com
I assume that that the Full Text engine will use the part of the Text that
is "in row" and the part that is not.
But when Full Text does a search:
join ContainsTable(position,*,'Computers')as ft on
p.PositionID=ft.[key]
I assume this is done from a dictionary somewhere and whether the Text field
is "in row" or not has no effect on this search. The only effect would be
when inserting the record or updating the record and changing the Text field
where the dictionary would be updated.
Are my assumptions correct?
Thanks,
Tom
> In general, if the bulk of your data is below the threshold where it goes
> in row, yes. Otherwise there isn't much of an improvement.
[quoted text clipped - 30 lines]
>>> >
>>> > Tom
Simon Sabin - 17 Oct 2006 00:38 GMT
Hello tshad,
I would store the data out of row.
Are you using SQL 2000 or sql 2005, if the later you can use varchar(max)
which is much better that text data types.
See my blog posts and article
http://sqlblogcasts.com/blogs/simons/archive/2006/10/04/SQL_Engine___Storage_of_
Large_Object_Data.aspx
Simon Sabin
SQL Server MVP
http://sqlblogcasts.com/blogs/simons
> I assume that that the Full Text engine will use the part of the Text
> that is "in row" and the part that is not.
[quoted text clipped - 86 lines]
>>>>>
>>>>> Tom