Lucene StoredField vs StringField performance

Question

In lucene indexing part, i started using StoredField, where i dont need to index the fields(before i was using StringField for the same fields).

This is basically for reducing some index space.

I am processing the file with 1 million records, but i dont see any performance difference. But soon i might have to process more than 100 million records.

so, i would like to know will there be any performance hit when i use StoredField compare to StringField?

Thanks Shankar

Answer 1

My guess is that your task is I/O-bound rather than CPU-bound as the most time is spent on document merging rather than simplified indexing (which, for StringField, is very cheap).

Profiling would likely help to reveal if this guess is incorrect.