I do not know if it's a bug or even implementation decision.
The fieldCache cache uses only the name of the field that was supposed
to be using the field name plus the name of the type, with that'd be
avoiding making unnecessary cache.
I had this problem and my memory consumption dropped from 80GB to 5GB.
Regardless of a different type, in Lucene it will end up under the same field name. What did you change? Are you loading for fields only for a specific type? This will generically cause more memory usage as you start to facet across types. I would say that it smells like you are using the multi types option wrongly?
I do not know if it's a bug or even implementation decision.
The fieldCache cache uses only the name of the field that was supposed
to be using the field name plus the name of the type, with that'd be
avoiding making unnecessary cache.
I had this problem and my memory consumption dropped from 80GB to 5GB.
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.