On Thu, Dec 4, 2008 at 11:41 AM, sunnyfr <[EMAIL PROTECTED]> wrote: > > Ok thanks a lot, > > so I can remove all this part
I wouldn't remove them if they are the source of a copyField (with the destination being text). Simply change to indexed="false" stored="false" otherwise you may get an undefined field exception. -Yonik > <field name="title" type="text" indexed="true" > stored="false" /> > <field name="description" type="text" indexed="true" > stored="false" /> > > <field name="tag1" type="string" indexed="true" > stored="false" omitNorms="true" /> > <field name="tag2" type="string" indexed="true" > stored="false" omitNorms="true" /> > <field name="tag3" type="string" indexed="true" > stored="false" omitNorms="true" /> > <field name="tag4" type="string" indexed="true" > stored="false" omitNorms="true" /> > <field name="tags" type="string" indexed="true" > stored="false" omitNorms="true" multiValued="true" termVectors="true" /> > > > and just keep ... : > <copyField source="title" dest="text"/> > <copyField source="title_en" dest="text"/> ..title_es ... > <copyField source="description" dest="text"/> > <copyField source="description_en" dest="text"/> ..title_en ... > <copyField source="tag1" dest="text"/> > <copyField source="tag2" dest="text"/> > <copyField source="tag3" dest="text"/> > <copyField source="tag4" dest="text"/> > > Just to be sure ... I index, title and description if for exemple i need to > boost them sepearetly... bf. title^2 description^1.5 > Languages one ... need to be index to apply analyzer/stemmer, and boost them > differently according to the country.. but I copy them to be searchable. > > thanks so much for your time .. again and again... > > > Yonik Seeley wrote: >> >> remove this entry from the example schema unless you need the >> timestamp when it was indexed: >> >> <field name="timestamp" type="date" indexed="true" stored="true" >> default="NOW" multiValued="false"/> >> >> Also, only index fields you really need to search separately. >> For example, if the "description" field is also indexed into the >> "text" field via copyField, and you only search it via the "text" >> field, then don't store or index the description field. >> >> Retrieving only ids is something that could be optimized in Solr, but >> hasn't been done yet. >> >> -Yonik >> >> On Thu, Dec 4, 2008 at 11:10 AM, sunnyfr <[EMAIL PROTECTED]> wrote: >>> >>> Hi Yonik, >>> >>> I will index my data again .... Can you advice me to optimize a lot my >>> data >>> and tell me if you see something very wrong or bad for the memory, >>> according >>> to the fact that I just need to show back the ID, that's it. >>> But I need to boost some field ... like description .. >>> description_country >>> according to the country ... >>> >>> Thanks a lot, I would appreciate it, >>> >>> >>> <fields> >>> <field name="id" type="sint" indexed="true" >>> stored="true" omitNorms="true" /> >>> >>> >>> <field name="status_private" type="boolean" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="status_... 6 more like that> >>> >>> <field name="duration" type="sint" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="created" type="date" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="modified" type="date" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="rating_binrate" type="sint" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="user_id" type="sint" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="country" type="string" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="language" type="string" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="creative_type" type="string" indexed="true" >>> stored="false" omitNorms="true" /> >>> >>> >>> <field name="rel_group_ids" type="sint" indexed="true" >>> stored="false" omitNorms="true" multiValued="true" /> >>> ... 3more like that >>> >>> >>> >>> <field name="rel_featured_user_ids" type="sint" indexed="true" >>> stored="false" omitNorms="true" multiValued="true" /> >>> <field name="rel_featured_group_ids" type="sint" indexed="true" >>> stored="false" omitNorms="true" multiValued="true" /> >>> >>> <field name="stat_views" type="sint" indexed="true" >>> stored="false" omitNorms="true" /> >>> ...6more like that ... >>> >>> <field name="title" type="text" >>> indexed="true" >>> stored="false" /> >>> <field name="title_fr" type="text_fr" >>> indexed="true" >>> stored="false" /> >>> <field name="title_en" type="text_en" >>> indexed="true" >>> stored="false" /> >>> <field name="title_de" type="text_de" >>> indexed="true" >>> stored="false" /> >>> <field name="title_es" type="text_es" >>> indexed="true" >>> stored="false" /> >>> <field name="title_ru" type="text_ru" >>> indexed="true" >>> stored="false" /> >>> <field name="title_pt" type="text_pt" >>> indexed="true" >>> stored="false" /> >>> <field name="title_nl" type="text_nl" >>> indexed="true" >>> stored="false" /> >>> <field name="title_el" type="text_el" >>> indexed="true" >>> stored="false" /> >>> <field name="title_ja" type="text_ja" >>> indexed="true" >>> stored="false" /> >>> <field name="title_it" type="text_it" >>> indexed="true" >>> stored="false" /> >>> >>> <field name="description" type="text" indexed="true" >>> stored="false" /> >>> <field name="description_fr" type="text_fr" >>> indexed="true" >>> stored="false" /> >>> <field name="description_en" type="text_en" >>> indexed="true" >>> stored="false" /> >>> <field name="description_de" type="text_de" >>> indexed="true" >>> stored="false" /> >>> <field name="description_es" type="text_es" >>> indexed="true" >>> stored="false" /> >>> <field name="description_ru" type="text_ru" >>> indexed="true" >>> stored="false" /> >>> <field name="description_pt" type="text_pt" >>> indexed="true" >>> stored="false" /> >>> <field name="description_nl" type="text_nl" >>> indexed="true" >>> stored="false" /> >>> <field name="description_el" type="text_el" >>> indexed="true" >>> stored="false" /> >>> <field name="description_ja" type="text_ja" >>> indexed="true" >>> stored="false" /> >>> <field name="description_it" type="text_it" >>> indexed="true" >>> stored="false" /> >>> >>> <field name="tag1" type="string" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="tag2" type="string" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="tag3" type="string" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="tag4" type="string" indexed="true" >>> stored="false" omitNorms="true" /> >>> <field name="tags" type="string" indexed="true" >>> stored="false" omitNorms="true" multiValued="true" termVectors="true" /> >>> <field name="owner_login" type="string" indexed="true" >>> stored="false" omitNorms="true" /> >>> >>> <field name="text" type="text" indexed="true" stored="false" >>> multiValued="false"/> >>> <field name="timestamp" type="date" indexed="true" stored="true" >>> default="NOW" multiValued="false"/> >>> <field name="spell" type="textSpell" indexed="true" stored="false" >>> multiValued="true"/> >>> <dynamicField name="random*" type="random" /> >>> >>> </fields> >>> >>> >>> >>> >>> Yonik Seeley wrote: >>>> >>>> On Thu, Dec 4, 2008 at 8:13 AM, sunnyfr <[EMAIL PROTECTED]> wrote: >>>>> >>>>> Hi Yonik, >>>>> >>>>> I've tried everything but it's doesn't change anything, I tried as well >>>>> the >>>>> last trunk version but nothing changed. >>>>> There is nothings that I can do about the indexation ...maybe I can >>>>> optimize >>>>> something before searching ??? >>>> >>>> Did you optimize the index (send in the optimize command) after >>>> indexing but before searching? >>>> >>>> curl http://localhost:8983/solr/update?optimize=true >>>> >>>>> I'm using linux system, apache 5.5, last solr version updated. >>>>> Memory : 8G Intel >>>>> >>>>> Do you think its a lot for the index size 7.6G for 8.5M of document? >>>> >>>> So it could be due to the index being slightly to big - subtract out >>>> memory for Solr and other stuff, and there's not enough left for >>>> everything to fully be cached by the OS. >>>> >>>> You can make it bigger or smaller depending on how you have the schema >>>> configured. >>>> The example schema isn't necessarily optimized for speed or size - it >>>> serves as an example of many field types and operations. >>>> >>>> Make sure you only index fields you need to search, sort, or facet on. >>>> Make sure you only store fields (marked as stored in the schema) that >>>> you really need returned in results. >>>> The example schema as copyFields and default values that you don't >>>> need - hopefully you've removed them. >>>> >>>> What's your schema, and do you have more examples of URLs you are >>>> sending to Solr (all the parameters)? >>>> >>>> -Yonik >>>> >>>> >>> >>> -- >>> View this message in context: >>> http://www.nabble.com/Solr-1.3---response-time-very-long-tp20795134p20836635.html >>> Sent from the Solr - User mailing list archive at Nabble.com. >>> >>> >> >> > > -- > View this message in context: > http://www.nabble.com/Solr-1.3---response-time-very-long-tp20795134p20837252.html > Sent from the Solr - User mailing list archive at Nabble.com. > >
