The purpose of the schema is to associate a type with a field name. That's it.
A dynamic field associates a type with a range of names. An empty field in a Lucene index doesn't take any space, so having 450 fields doesn't in itself cause a problem. The point at which you may have a problem is when you want to search across those dynamic fields, as search across multiple fields in Solr/Lucene is not as efficient as searching across a single field. If in your application you will know the name of the field you want to search against at query time, then your scenario seems quite reasonable. Upayavira On Fri, Mar 15, 2013, at 09:51 AM, kobe.free.wo...@gmail.com wrote: > Hello All, > > Scenario: > > We trying to define the schema structure for our application search > feature, > based on SOLR search server. In our scenario the total number of fields > is > 450 (quiet huge) and we will be using the features like faceting, sorting > etc. This field set will be dynamic (not permanent) and will be modified > (like removing/ adding some fields) on regular basis, based on the > business > needs. We are planning to use the concept of 'Dynamic Fields" for most of > the fields from the original fields set of 450. > > Queries: > > 1. What are the pros/ cons of using dynamic fields? > 2. What is the best way to achieve dynamic schema, is the above mentioned > method proper? > 3. What will be the best approach to update the schema and re-index the > data? > 4. Is there a major index size difference when using dynamic fields? > > > > > > -- > View this message in context: > http://lucene.472066.n3.nabble.com/SOLR-Dynamic-Schema-Design-tp4047638.html > Sent from the Solr - User mailing list archive at Nabble.com.
