We’re using MongoDB Atlas, and we’re already using ElasticSearch.
However, we face a lot of problems with the added burden of syncing data and schema changes between MongoDB and ElasticSearch, plus the extra cost of maintaining an ElasticSearch self-hosted instance.
We’re seeking the best full text-search across multiple collections, with suggestions and auto-complete functionality.
We are now considering MongoDB Atlas Search.
Here are the main issues we face with Atlas Search:
1 - Atlas Search requires Searching to happen as the first stage in an aggregation pipeline. This puts a limitation because sometimes we want to do an aggregation first (lookups, match, etc) then do the search on the data from the previous stages. Does anyone know if this is still a limitation or has things changed ? And what is the best approach in this case ? i.e. How can we search across multiple collections for example using the same search query ?
2 - Does Atlas Search provide the same auto-complete functionality as ElasticSearch ?
3 - Does Atlas Search provide the same search-suggestions functionality as ElasticSearch ?
Many thanks for your time and support