My experience so far is that tuning the configuration is a bit of a mystic art. I have not had much success trying to figure out how to improve the index performance (ES5 Indexing performance (seems slow)).
Threadpools are stored in memory, by dramatically increasing the size you risk OOM. Also 12K requests might take a while to process, what do your clients do in the meantime?
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.