In Lucene, what is the difference between ANALYZED and ANALYZED_NO_NORMS?

Question

I could not understand the difference between the two indexing methods: ANALYZED and ANALYZED_NO_NORMS . I read Lucene Javadoc but didn't understand the difference.

Can someone tell me more about NORMS? What are the benefits or limitations that they bring to indexing?

Answer 1

analyzed

Index tokens created by running the field value through the analyzer. This is useful for general text. The analyzer can be something like a snow cat stalk analyzer:

• http://e-mats.org/2009/05/modifying-a-lucene-snowball-stemmer/

ANALYZED_NO_NORMS

It uses an analyzer, but it does not create norms for fields.

• http://lucene.apache.org/java/2_4_0/scoring.html

Norms are created to quickly count documents at the time of the request. Typically, these norms are loaded into memory, so when you run the query analyzer by index, it can quickly evaluate search results.

No norms mean that the index field and document elevation and normalization of the field length are disabled. The advantage is that memory usage is less, because in each search by index during the search, the norms take one byte of RAM per indexed field for each document in the index.