-
Notifications
You must be signed in to change notification settings - Fork 1.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Support Azure Cosmos for NoSQL, SQL API #2713
Comments
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
@kylehayes @jferrettiboke: Thanks for reaching out and Cosmos is definitely on our mind! I have 2 questions for you that would help us greatly in setting priorities:
|
I thought Cosmos was multi model? Whichever flavour supports serverless would get my vote! |
That’s a great question. Of the teams I know using it, it seems split down the middle between the SQL and Mongo flavor. It does support other models as well like a graph db but not sure how popular those are. The SQL one is more well supported in the Azure ecosystem but the Mongo flavor I can see people choosing to be Mongo compatible. My vote would be SQL. This is relatively moot for our team as we are currently converting from Cosmos to Postgres and switching our custom ORM over to Prisma. |
My company has been using Cosmos as our main database for a bit now. We worked directly with Microsoft for configuration. Serverless is a terrible idea to support. It is in "Preview" and has massive restrictions around it's scalability and no planned release date. The better one to support would be Autoscale as it has most of the features (though not all) of Serverless, much more support from MS, and has a stable release. MS also recommends using the SQL version of the API as the mongo one is significantly less featureful. Though it was recently updated to version 4.0 |
SQL API is considered the "main" CosmoDB API; Unfortunately its also the API with the most restrictions I.E no createMany, CosmoDB does not allow any kind of lookups based on the result of a sub query or cross document joins this restriction is quite painful to have to work around. I'd love Prisma for the "main" CosmoDB SQL API is this is the API we use on our IoT project, yes prisma will essentially launch defacto support for the MongoDB CosmoDB API Soon, however people on the SQL API can't migrate to the MongoDB API in a non-trivial way, the containers would have to be re-created from scratch and data migrated over aswell as "id" fields mapped to their CosmoDB equivalent. |
Just decided to build this myself, already most of the way through will post a npm link here when its done but this is the high level plan which could technically be replicated out to other databases.
The gif is a proof of concept ignore "distinct" not being added to the query that's on the way but nested and/or's work |
If you let me express my opinion, I need support for SQL as well. The project I am working on is a very good example of this necessity. For now, I am using MongoDB, however, we're are keen to step out from it to another database, on Cosmos the issue would be kind of solved, but for the code itself I still wish using Prisma as my repository layer, which makes life easier to migrate from DBMS, given I just need to change the services that uses it and easily migrate the DBs. |
Can someone provide current documentation links for the Cosmos DB, SQL API? |
@janpio it's called |
Thank you. That naming really takes some effort to wrap your head around. But indeed:
https://learn.microsoft.com/en-us/azure/cosmos-db/choose-api#coresql-api And then: https://learn.microsoft.com/en-us/azure/cosmos-db/nosql/query/ |
You can also use free tier for the dev purposes > https://learn.microsoft.com/en-us/azure/cosmos-db/free-tier (one per subscription) or serverless which is charging per usage (without monthly fee) |
It would be awesome to have support for Azure Cosmos DB.
Read this for more info and the benefits of using Azure Cosmos DB.
Discussion: #2704
The text was updated successfully, but these errors were encountered: