If we want to index lots of data of the form <Key, Value> (all of the data have the same structure. For example, all of the keys contain values for the same two fields), and then just query them (lookup queries and not e.g., range queries) a lot without any inserts/updates/deletes, is there any reason to prefer elasticsearch over a relational database with an index on the key columns? My main consideration is how fast they can answer the lookups.
If you are not looking for fulltext search (as like %term%), that you don't care about relevancy and the total number of elements to store does not require to scale out (multiple physical machines) then you can consider not using elasticsearch.
My recommendation in general though is to use elasticsearch anytime you have a search use case or an analytic use case (or both).
Disclaimer: as an elastic employee for a long time, I might be biased now 
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.