You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-user@lucene.apache.org by wolvernie88 <mw...@gmail.com> on 2008/05/09 17:38:43 UTC

Is this the right way to use Lucene in multithread env?

Hi,

Here is what I am using Lucene.

I build the index (from different data source) during midnight. I build a
FSDirectory. Then I load it into RAMDirectory for the best performance. When
I built it, I called IndexWriter.optimize() once. 

Once the index is built, I will never update it.

I have static variable defined as IndexSearcher. Once I load RAMDirectory, I
do

		        newIndexDirectory = new RAMDirectory(fsDirectory);
			IndexWriter newWriter = new IndexWriter(newIndexDirectory, new
StandardAnalyzer(), true);
			newWriter.optimize();
			newWriter.close();
		        searcher = new IndexSearcher(newIndexDirectory );

For every new search, I do

			QueryParser parser = new QueryParser("field1", new StandardAnalyzer());
			Query query = parser.parse(queryString);
			Hits hits = searcher.search(query);

Is this the right way? Do I need to close parse, query or hits?

As I have only one IndexSearcher, will it cause any problem?

I found  using the same query does not always give me the same response
time.

Thanks much.
-- 
View this message in context: http://www.nabble.com/Is-this-the-right-way-to-use-Lucene-in-multithread-env--tp17150728p17150728.html
Sent from the Lucene - Java Users mailing list archive at Nabble.com.


---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org

Re: Is this the right way to use Lucene in multithread env?

Posted by wolvernie88 <mw...@gmail.com>.
Hi,

I am creating a new indexWriter to optimize the directory.

I will try using FSDirectory later.



Otis Gospodnetic wrote:
> 
> Hi,
> 
> No need to close parse and it is good to use the same searcher.
> I don't understand why you have that IndexWriter there if you are
> searching...
> 
> Also, you may not benefit from explicit loading of the index into RAM. 
> Try without it first.
> 
> 
> Otis
> --
> Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch
> 
> 
> ----- Original Message ----
>> From: wolvernie88 <mw...@gmail.com>
>> To: java-user@lucene.apache.org
>> Sent: Friday, May 9, 2008 11:38:43 AM
>> Subject: Is this the right way to use Lucene in multithread env?
>> 
>> 
>> Hi,
>> 
>> Here is what I am using Lucene.
>> 
>> I build the index (from different data source) during midnight. I build a
>> FSDirectory. Then I load it into RAMDirectory for the best performance.
>> When
>> I built it, I called IndexWriter.optimize() once. 
>> 
>> Once the index is built, I will never update it.
>> 
>> I have static variable defined as IndexSearcher. Once I load
>> RAMDirectory, I
>> do
>> 
>>                 newIndexDirectory = new RAMDirectory(fsDirectory);
>>             IndexWriter newWriter = new IndexWriter(newIndexDirectory,
>> new
>> StandardAnalyzer(), true);
>>             newWriter.optimize();
>>             newWriter.close();
>>                 searcher = new IndexSearcher(newIndexDirectory );
>> 
>> For every new search, I do
>> 
>>             QueryParser parser = new QueryParser("field1", new 
>> StandardAnalyzer());
>>             Query query = parser.parse(queryString);
>>             Hits hits = searcher.search(query);
>> 
>> Is this the right way? Do I need to close parse, query or hits?
>> 
>> As I have only one IndexSearcher, will it cause any problem?
>> 
>> I found  using the same query does not always give me the same response
>> time.
>> 
>> Thanks much.
>> -- 
>> View this message in context: 
>> http://www.nabble.com/Is-this-the-right-way-to-use-Lucene-in-multithread-env--tp17150728p17150728.html
>> Sent from the Lucene - Java Users mailing list archive at Nabble.com.
>> 
>> 
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
>> For additional commands, e-mail: java-user-help@lucene.apache.org
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-user-help@lucene.apache.org
> 
> 
> 

-- 
View this message in context: http://www.nabble.com/Is-this-the-right-way-to-use-Lucene-in-multithread-env--tp17150728p17152621.html
Sent from the Lucene - Java Users mailing list archive at Nabble.com.


---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org