-
Aggregated data is not likely to be stored in the same shape as it is in a Rdbms
-
Rdbms more likely to store data in related tables using foreign key relationship, whether enforced or not
-
Aggregating data in ElasticSearch from Rdbms is likely to result in data de-normalization
-
Rdbms likely data model:
-
Movies
AssetId | Title | ShortSynopsis | ... |
---|---|---|---|
86c1bba8-d18f-4bbc-9cb4-a90a4220f59c | My First Mister | Desperate to escape the.. | ... |
- Cast Members
MemberId | Name | ... |
---|---|---|
b42e5484-e7e1-4eaf-b9a2-b8f25487533b | Leelee Sobieski | ... |
1290de91-2bc3-4b72-9a0d-6fa0e04b44ab | Albert Brooks | ... |
- Movie Cast Members
AssetId | MemberId | Role |
---|---|---|
86c1bba8-d18f-4bbc-9cb4-a90a4220f59c | b42e5484-e7e1-4eaf-b9a2-b8f25487533b | Cast |
86c1bba8-d18f-4bbc-9cb4-a90a4220f59c | 1290de91-2bc3-4b72-9a0d-6fa0e04b44ab | Cast |
- Document databases such as Couchbase and MongoDb may store data aggregated neatly already