On: How we built a serverless SQL database

[rabaut]

Does CockroachDB Serverless expose an HTTP api? This sounds like a great fit for use with Cloudflare Workers, but that requires an http api.

[andydb → rabaut]

Great question. We recognize how important this is and are actively working on it.

[webmaven → andydb]

Is it possible to use PostgREST in front of CRDB to solve this use case?: https://postgrest.org/

[andydb → webmaven]

We're working on enabling PostgREST. There is some small-ish functionality that we need do in CockroachDB for it to fully work.

[sorenbs → rabaut]

Prisma will launch a data proxy, as well as support for cockroachdb and Cloudflare workers early next year.

[password4321 → rabaut]

I think Supabase does, 500mb storage + 2gb outbound data transfer for free.

[sharps_xp → rabaut]

Why do you prefer an HTTP API versus a DB connection? Isn't the former going to inherently have the overhead cost of creating the connection + TLS handshakes?

My question is similar, which is, is CockroachDB going to have an equivalent RDS proxy so that apps can handle traffic spikes and not have to deal with problems with DB connection pools

[pistoriusp → sharps_xp]

I think we won't be getting socket connections in some of the wasm powered JS runtime engines soon. Using http solves that, and a bunch of caching issues.

[estambar → pistoriusp]

there are some great projects that present an http api for postgres that you could use with CRDB I think. I'm thinking of something like https://github.com/pramsey/pgsql-http

[pistoriusp → estambar]

Now it would be awesome if we could use Prisma against an open standard like this.

[zaidf → rabaut]

I kind of assumed this came with an HTTP api, woops :) I've been using Airtable with Workers for tiny side projects. This seemed like a nice alternative to Airtable's lack of support for vanilla SQL.