On Wed, Aug 19, 2009 at 6:44 PM, ahammad <ahmed.ham...@gmail.com> wrote:
> > Each core has a different database as a datasource, which means that they > have different DB structures and fields. That is why the schemas are > different. > > If all the shards should have the same schema, then what is the point of > > sharding in the first place? I thought that it was used to combine > > different cores with different index structures...Right now, every core I > > have is unique, and every schema is different... > > > Index is sharded when it becomes too much for one box to keep the whole index. Distributed Search in Solr can merge these multiple indexes running on different boxes into one result set. It is not meant for combining different cores or different schemas. If many shards have a document with the same uniqueKey value, any one can be returned. Typically, shards have the same schema, with each having a disjoint subset of the complete set of documents. -- Regards, Shalin Shekhar Mangar.
