I found this comment in the elasticsearch6.1.3‘s source code about index.store.preload
/** On which extensions to load data into the file-system cache upon opening of files.
* This only works with the mmap directory, and even in that case is still
* best-effort only. */
my index.store.type is niofs, maybe I mismatched this parameter.
I guess that all the terms are quite common which should result in a high nextdoc count in the profiler output which would explain what is taking so long.
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.