I need to get all documents from index, so for that I using scroll to get all documents in bulks. Since my goal is to perform as less round-trips to elastic as possible, I want bulk size as big as possible.
What is the maximal size of scroll window? Sorry, can't find this number in documentation...
Sure you can modify the default maximum value but I hope you understand the consequence of it and the memory pressure this is going to add on your nodes as nodes will have to send more data to the coordinating node which will have to hold that in memory.
I'd recommend doing tests of this solution.
It's a tradeoff between round-trips and memory usage.
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.