The default implementation of google guava cache is in-memory. Does
elasticsearch override this default behavoir?
If not, please some one tell me how to make cache persistable probably to
index reposioty itself when the cached values are complex objects
There are several caches in Elasticsearch, the main ones being the filter
cache and the field data cache. Storing the filter cache on disk makes
little sense because in most cases it would make the cached filters almost
as slow to load as the original filters. Regarding field data, it is not
possible to have disk-based field data implementations in current releases
but there is ongoing work towards that direction so this feature will
hopefully be available in a future elasticsearch release.
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.