I would say it all depends on what you are trying to do. Unlike a relational database, in Solr the data does not need to be normalised, you need to put everything into an index so that you can achieve whatever feature it is that you want. For example, you may search on customer and want a facetted count of the products.
Also in Solr you have the concept of multi valued fields, therefore you could have a product index with a multi valued field that stores say customer id, thereby linking products and customers in one index. We have multiple cores which we had to create for various reasons. To access the cores for indexing (or searching) you just have to refer to the cores by their names in any Solr URLs or in the Java client etc. I think it all depends on what it is you are trying to achieve. It's best to read a good book such as http://www.packtpub.com/solr-1-4-enterprise-search-server/book and see what can be achieved and design your indexes accordingly. Hope that helps. On 20 June 2011 05:38, jboy79 <joel_pangani...@yahoo.com> wrote: > Hi, I am new to SOLR and would like to know if multiple cores is the best > way > to deal with having a product and customer index. If this is the case how > do > you go about indexing on multiple cores. > Thanks > > -- > View this message in context: > http://lucene.472066.n3.nabble.com/Multiple-Cores-tp3084817p3084817.html > Sent from the Solr - User mailing list archive at Nabble.com. >
