Recently, i found that the mapping types will be removed once v6.0 is released.
I would like to know the thoughts on suggested alternatives:
Create index per type - I have around 30types. So indexes created will be 30. Does ES perform better while searching multiple indices ? In the past i have been reading posts where people suggest on keeping the number to indices to minimum for better search.
Single index, but a "type" field to differentiate the type - It would be a bit difficult to visualize data esp. in HEAD plugin or any like software. Will this approach perform better while search than the #1 above ?
Misc query:
What will be the impact on indexing performance on both above alternatives ?
Thanks Mark. My documents are almost similar for the types. Just a field here and there.
I have a case perfectly suited for Nested type. But due to indexing performance issues & searching simplicity, we have created different types and the data is repeated for documents.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.