We have requirement where we first need to add the document and then update the document.
We are for now, using update api to do add and update.
I see many reads in disk and if we disable the indexing then reads decrease to great extend.
Is update api is making the read?
If I change the update api to index api with ID, would it make any difference.
I have this question as I see if you pass ID in index api, and try to index same document twice with same id, then Es would change the version of document, Which made me suspicious that, ES is doing read is index as well. Is that true?
Would changing from UPDATE to Index make any difference on READ?
Index a document is a write (index operation of the whole content).
Index again a document is a delete + index operation of the whole content
Update a document is a get + merge in memory the previous content with the new one + delete + index operation of the whole content
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.