I was looking in using elasticsearch as my desired database as it is very impressive architecture and performing system. One thing i noticed was .. for every index updated, there is a version being created. If not mistaken, it dose stores the older versioned objects too.. as elastic indexes seems to be immutable. Now in such a scenario, the data space is going to exponentially explode in case of me having a scenario .. where i constantly update a certain set of values.
Correct me if i am wrong in my interpretation about my system.
I wanted to understand . in such a scenario - will it be a good option to have elasticsearch as an option to go for or will you recommend me some other solution.
indeed Elasticsearch is an append only datastore. Once a document is written to disk, it is immutable. However there is a regular clean up process going on called merging, which rewrites your data and is removing all the documents that are marked as deleted.
Also, Elasticsearch does not keep revisions of documents around. You can access the latest version and that's it. Others will be removed over time due to merging.
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.