Concerns about Dynamic Field Mapping

I'm new to Elastic Search and I read about the feature of dynamic field mapping at

It's not possible to change an exist field mapping unless you take the REINDEX action. But what will ES do when we just add a new field mapping ( or index a document with new properties)? is this a heavy work for ES when many documents already exist in this index?

Assuming we have 10,000,000 documents in "INDEX_1". Each of them has a similar structure like:
"name" : xxx,
"age": xxx

Now we index a new document:
"name": "Kevin",
"age": 22,
"new field" : "something"

I wonder whether the index action of the new document will affect the existing 10,000,000 document and have a great impact on cluster performance.


This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.