Re: PDFs in blobs and SQL2005
- From: Simon Sabin <SimonSabin@xxxxxxxxxxxxxxx>
- Date: Wed, 19 Jul 2006 22:44:57 +0000 (UTC)
To get round this you need to pass the date criteria into the field indexed by FT. FOr this to work you need to store the date in your serach field. This is more difficult with PDFs. you may have to convert to text and then store and index that.
You can then do a search like this
select * from tablename where contains(*,'test') and
datecol between '1/1/2005' and '2/1/2005'
select * from tablename where contains(*,'test and (DT20050101 OR DT20050102)')
They key to getting a performance FT system is to restrcit the data that comes back from the FT engine
Simon Sabin
SQL Server MVP
http://sqlblogcasts.com/blogs/simons
something like this
select * from tablename where contains(*,'test') and
datecol>getdate()-365
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
"DoctorN" <ron at stringx.com> wrote in message
news:uOeFBQeqGHA.3484@xxxxxxxxxxxxxxxxxxxxxxx
Hi
Thanks for taking the time to reply Hilary
I understand what you say, but could you please clarify " and are not
passing them on for further restrictions, i.e. date." ?
Your help appreciated
"Hilary Cotter" <hilary.cotter@xxxxxxxxx> wrote in message
news:uSNdy7TqGHA.2464@xxxxxxxxxxxxxxxxxxxxxxx
Its possible depending on the complexity of your queries and if you
can limit them to a small number of results and are not passing them
on for further restrictions, i.e. date.
--
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
"DoctorN" <ron at stringx.com> wrote in message
news:eoOl0lPqGHA.4408@xxxxxxxxxxxxxxxxxxxxxxx
Hi there,
I've been playing around with full text searching and the PDF
ifilter and all seems good.
I have a possible project that involves indexing nearly 100 million
1-4 page pdf files and requires a full text search. I realise this
is significant and have 2x 8 core 8gb ram servers at my disposal
and terrabytes of space
The client expectations are to have searches return within 10
seconds maximum using full text and i wonder if any of you have had
similar projects to which i can gain advice and avoid pitfalls. I
am familiar with query tuning but not that familiar with FT
searching as yet. Also the PDFs will be in blobs internally.
Any thoughts on whether this search peformance is any where
achievable? ps, there may be up to 400 users on the system, though
it's not clear how concurrent.
.
- Follow-Ups:
- Re: PDFs in blobs and SQL2005
- From: Hilary Cotter
- Re: PDFs in blobs and SQL2005
- Prev by Date: Re: FTS Performance in SQL 2005
- Next by Date: Re: change tracking, update in background not working
- Previous by thread: change tracking, update in background not working
- Next by thread: Re: PDFs in blobs and SQL2005
- Index(es):
Relevant Pages
|