Thanks, I am trying your method. And do you think 460ms is a long time for my computer(2 cores, 2GB memory)?
Regards, Rong Kang At 2012-02-09 18:37:15,"Rafał Kuć" <r....@solr.pl> wrote: >Hello! > >I don't know what Your queries will look like, but let's assume that >you will use 3 fields for searching like title, description and >author, you will also filter on category field and sort on title_sort >field, than you could use a warming query like the following: > ><arr name="queries"> > <lst> > <str name="q">common query</str> > <str name="qf">title description author</str> > <str name="fq">category:books</str> > <str name="sort">title_sort asc</str> > <str name="defType">dismax</str> > </lst> ></arr> > >Please consider this only an example. You need to look at Your >application and choose common queries and common query parameters like >sorting, filtering and such. > >-- >Regards, > Rafał Kuć > Sematext :: http://sematext.com/ :: Solr - Lucene - Nutch > >> Hi! > > >> Thanks for your reply. I read some material at >> http://wiki.apache.org/solr/SolrConfigXml#newSearcher , but it doesn't give >> me some example. >> Could you tell me how to modify it or give some links about incorporating >> warming queries ? >> > >> Regards , >> Rong Kang > > >> At 2012-02-09 18:01:38,"Rafał Kuć" <r....@solr.pl> wrote: >>>Hello! >>> >>>You may want to incorporate warming queries to your solrconfig.xml >>>file. Edit the solrconfig.xml file and look for the default warming >>>queries, they are in the following places: >>> >>><listener event="newSearcher" class="solr.QuerySenderListener"> >>>and >>><listener event="firstSearcher" class="solr.QuerySenderListener"> >>> >>>Modify the default ones to match your needs - for example include >>>sorting there, fields you use for searching, filtering and etc. >>> >>>-- >>>Regards, >>> Rafał Kuć >>> Sematext :: http://sematext.com/ :: Solr - Lucene - Nutch >>> >>>> My solr 's index data size is 1.1GB,and solr only has one core. >>> >>> >>>> When I first search one word in solr . its response time is 460ms. >>>> When I search the same word the second time. its response time is >>>> under 70ms. I can't tolerate 460ms . Does anyone know how to improve >>>> performance? >>> >>> >>>> My Computer: >>>> 1 cpu(with 2 core) >>>> 2G memory (800MHz) >>> >>> >>>> and my search configuration >>>> <str name="defType">dismax</str> >>>> <str name="qf">filename^5.0 text^1.5</str> >>> >>> >>>> <str name="q.alt">*:*</str> >>>> <str name="hl">on</str> >>>> <str name="hl.fl">filename text</str> >>>> <bool name="hl.useFastVectorHighlighter">true</bool> >>>> <str name="hl.tag.pre"><![CDATA[<b style="color:red">]]></str> >>>> <str name="hl.tag.post"><![CDATA[</b>]]></str> >>>> <int name="hl.fragsize">100</int> >>>> <str name="f.filename.hl.alternateField">filename</str> >>>> <int name="f.text.hl.snippets">3</int> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> > > > >
