M0 is MongoDB Atlas’s no-cost, shared cluster tier. It provides a replica set for learning MongoDB, trying features, and building small, low-throughput development projects. Shared hardware keeps costs down but limits resources, performance, and feature access. section 10.1
Check these limits before choosing M0:
- Each Atlas project can have one M0 cluster. Its memory and storage size cannot be configured, its three-node replication factor is fixed, and it cannot be deployed as a sharded cluster. section 10.1
- M0 has no Atlas Cloud Backup. Use mongodump to make a logical backup and mongorestore to recover it. Atlas does not support their oplog options for point-in-time capture or replay. section 21.1.1 section 21.1.2
- Do not use M0 for production workloads or sensitive data: shared tiers lack advanced security features and have operational limits. section 10.1
When the project outgrows M0, use Edit Configuration in the Atlas UI and choose Flex or M10 under Cluster Tier. Allow for 7–10 minutes of downtime when changing among these tiers. section 10.1
Try vector search on M0
M0 supports Atlas Vector Search. List the cluster with `atlas cluster list`, create a search index from a definition file, then check its status with `atlas clusters search indexes list --clusterName MongoDB-in-Action --db sample_mflix --collection embedded_movies`. section 12.2.4
atlas cluster search index create \ --clusterName MongoDB-in-Action \ --file vector-definition.json
This Atlas CLI command creates the example vector search index from `vector-definition.json` on an M0 cluster. section 12.2.4
