Best compression will compress source more efficiently, and should save you some space. If you are using the default mappings you can optimise these depending on what you know about your data. This blog post covers this, and even if it is getting a bit old most of it is still valid. Compression efficiency also tend to at least to some extent depend on shard size, so if you have lots of small shards, you may get better compression rates by consolidating them.
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.