More generally, if you have two concurrent indexing requests then it requires coordination to put them into a specific order, and this can be quite time-consuming. Elasticsearch tries to avoid doing this kind of cross-request coordination as much as possible.
If you want to put all your documents into a specific order you will need to do so yourself. If you can tolerate a small amount of disorder then perhaps you can just use timestamps.
Thank you for the quick reply.
Regarding timestamp, if i understand correctly, we need to have a custom timestamp field, and sort based on timestamp. correct?
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.