Introducing DocumentDB 1.0: our 19th Cloud engine
Today we are adding DocumentDB 1.0 to Layerbase Cloud, our 19th hosted database engine. It joins FerretDB as another choice for applications that use MongoDB drivers and want an alternative to Atlas.
Create a DocumentDB database, or compare our MongoDB-compatible engines.
Another choice for MongoDB workloads
We have supported FerretDB for a long time. Some Atlas users have made the move, but we have also heard hesitation. Choosing a database is a long-term decision, and the uncertainty around MongoDB's lawsuit against FerretDB, filed in May 2025, has been part of those conversations. That is not a judgment on the merits of the case or on FerretDB's technology. It is a reason we want customers to have more choices.
We are excited to see DocumentDB reach its first generally available release. Originally developed at Microsoft and now a Linux Foundation project, DocumentDB is MIT licensed. It stores BSON documents using PostgreSQL and exposes a MongoDB-compatible gateway.
This is the standalone open-source DocumentDB project, not Amazon DocumentDB or the managed Azure service. FerretDB remains supported on Layerbase. Existing databases are not converted, and nothing about their connection details or data changes with this release.
Use familiar drivers, test your actual workload
DocumentDB speaks the MongoDB wire protocol. You can connect with mongosh or a MongoDB driver using the connection string from your Layerbase dashboard. For example, after connecting to a disposable database:
db.tasks.insertOne({
title: 'Try DocumentDB on Layerbase',
status: 'ready',
owner: 'developer'
})
db.tasks.find({ status: 'ready' }).toArray()That makes it a useful option for document-oriented application data, prototypes and development environments. Compatibility is not a promise that every MongoDB or Atlas feature behaves identically. Test your queries, indexes, aggregation pipelines and authentication requirements before moving a production workload. Our connection guide covers getting connected.
The guided Atlas migration currently imports into FerretDB. For DocumentDB, use a separately tested export/import workflow and validate the result in a disposable database before cutover. Keep the source until you have checked representative documents, indexes and application behavior. The two engines have separate compatibility profiles; FerretDB benchmarks are not DocumentDB benchmarks.
Branching matters more as agents write more code
Atlas users have told us they want an easier way to give each preview environment or coding agent its own isolated database. We agree. An agent testing a migration or rewriting data should have a disposable environment, not a production connection string.
DocumentDB supports database branching on Layerbase, alongside FerretDB. Create an isolated branch for an agent, preview deployment, or test; writes to that branch stay separate from the parent. Reset the branch from its parent when you need a fresh copy. See the branching guide for the workflow and supported engines.
Solo and Pro, with a free way to try it
DocumentDB is available on Solo at $5/month and Pro at $15/month. It is not on the Free plan at launch. A fresh instance has a modest footprint, but we need more testing of concurrent queries, indexing and restores within the Free plan's memory budget before offering it there.
You can try DocumentDB with the seven-day Pro free trial. A payment method is required; the trial converts to $15/month unless canceled. See the current terms on the pricing page.
Pro includes up to 10 databases, 25 GB of storage and a shared 1.5 GB always-on RAM pool. Ingress and egress have no per-GB bill. They are included subject to the acceptable-use policy: combined traffic across your databases is limited to 100 GB per rolling day, 500 GB per rolling week and 1,500 GB per rolling 30 days. Those bandwidth allowances are separate from memory and storage limits. We think that is generous room to build without a surprise transfer bill.
Free local development, without Docker
On Linux x64 and ARM64, you can run DocumentDB locally for free through the Layerbase CLI, without Docker or a Cloud subscription. With a current installation:
npm install -g layerbase
lbase create documentdb-demo --engine documentdb --db-version 1.0.0 --start
lbase connect documentdb-demoThe CLI uses the same native runtime tooling behind our hosted integration. We are working on native Mac and Windows integration. DocumentDB support in Layerbase Desktop is also planned; the current Desktop release does not yet include it. Linux CLI availability should not be read as availability on every local platform.
Get started
Create DocumentDB on Layerbase, connect your application and try the queries that matter to you. If you are choosing between DocumentDB and FerretDB, the MongoDB-compatible engine comparison explains the distinction before you create anything.
Our Cloud catalog now has 19 engines. DocumentDB gives you another implementation to evaluate, with a flat plan price and the option to develop locally on Linux before deciding where to host.
Keep reading
- MongoDB CEO transition: CJ Desai leaves, Dev Ittycheria returns as interim CEOMongoDB announced an immediate CEO transition on September 28, 2026. Here are the confirmed facts and the practical questions for teams running MongoDB.
- Northflank alternatives: a preview environment is not a database branchNorthflank gives every pull request its own stack, and the database in that stack starts empty unless you seed it or restore it from a backup you already had. Here is exactly how their forks work, what a copy-on-write branch does differently, what each one costs while a PR sits open, and the cases where Northflank is the right answer.
- MongoDB Atlas alternatives: flat pricing and a document database you can branchWhat you actually get when you leave MongoDB Atlas for managed FerretDB on Layerbase: the same wire protocol and driver, a flat monthly price instead of a per-tier meter, branching Atlas does not offer, and an honest list of the cases where you should stay put.
- MongoDB vs FerretDB on the same hardware: the benchmark, and what we had to fix to run it fairlyYCSB against MongoDB 8.0 and FerretDB 2.7 on the exact hardware behind our $65/mo dedicated preset. Once our own tuning was right, FerretDB held 3-4x MongoDB throughput on the mixed workload and about 5x on reads. The corrected numbers, what had to be fixed first, and what we are not claiming.