lucenenet-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "George Aroush" <geo...@aroush.net>
Subject RE: Keep DotLucene indexes Safe
Date Mon, 03 Jul 2006 20:42:52 GMT
Hi Vincent,

I have posted your question to Lucene.Net mailing list.  Please subscribe to
the mailing list so you can follow up on your question and post on your own.
Please see: http://incubator.apache.org/projects/lucene.net.html on how to
subscribe so you can post.

Regards,

-- George Aroush 

-----Original Message-----
From: Vincent Daron [mailto:vdaron@ask.be] 
Sent: Friday, June 30, 2006 5:26 AM
To: lucene-net-user@incubator.apache.org
Subject: Keep DotLucene indexes Safe

Hi

I'm doing some tests using DotLucene (with Wikipedia data) and I have some
question about Lucene.NET.

1) I'm using multiple threads to index documents and one thread to write
RamDirectory indexes on disk ( as only one writer is allowed). Bu using that
mechanism and the MergeIndexes methods, an Optimize is done before each
merge.That optimize remove all interest of creating Memory indexes.
Right ? It seems much faster to add Document sequencially instead of Merging
RAMDirectory indexes.

2) Can an Optimise corrupt an index ?
I've ran into that problem using Luke to optimise an index (about 2Gb never
optimised). Once the Optimise was done, my index was corrupted.
Any idea ?

Thanks a lot for your time and your work.

Vincent



Mime
View raw message