What is the simplest tool to do the job? Or maybe better yet: Use something that's already (or will be), in your technology stack to avoid having one more thing in it.
What is the simplest tool to do the job? Or maybe better yet: Use something that's already (or will be), in your technology stack to avoid having one more thing in it.
> What is the simplest tool to do the job?
For me it's definitely MongoDB but I'm afraid of data loss etc. My example was simplified. In reality I want to update a sub-array within the collection, like this: collection['some_identifier']['items'][3] = 123. Overwriting the whole 'some_identifier' document seems wasteful. With MongoDB it's easy though to update nested sub-data.
When you need anything more (which is probably "soon" if not "already"), it'll bite one way or the other.
If you batch operations it is extremely fast at updates across multiple documents.
Compare aggregation framework to what SQL has, for example:
https://docs.mongodb.com/manual/reference/sql-aggregation-co...
Or even within Mongo, querying and aggregating are astonishingly different things, that you need to learn separately.
And then there are JOINS. And stale reads with other serialization issues.