More fields will cause some increase in memory use, but it's hard to quantify it, and it will also depend on how much usage the queries can make of Solr's caches, number of simultaneous queries. In general, the number of query fields has more impact on performance.
A potentially big memory sink, which will depend on the number of fields in use, is in the Lucene FieldCache, which is used when result sets are to be sorted. If you haven't a;ready done so, take a look at http://wiki.apache.org/solr/SolrPerformanceFactors which describes a lot of the factors influencing performance and memory use. -Simon On Fri, Sep 30, 2011 at 8:27 AM, Pranav Prakash <pra...@gmail.com> wrote: > How will the number of fields increase the amount of RAM usage in Solr 3.4? > I have about 37 different fields because i've made every field for every > language. I shall be doing dismax search all the time (or slightly less > than > all). So, I will have to send my query like > > *qf=title_en,title_ar,title_hi,title_fr... and so on. > > Will it cause significantly high memory usage than had there been a unified > field with no langauge specific tokenizers? > > *Pranav Prakash* > > "temet nosce" > > Twitter <http://twitter.com/pranavprakash> | Blog <http://blog.myblive.com> > | > Google <http://www.google.com/profiles/pranny> >
