MongoDB DBA Associate key terms
MongoDB DBA Associate · Get comfortable with the vocabulary.
- replica set
- Related: oplog, primary, election
A group of mongod processes that maintain the same data set. One member is primary and accepts writes; secondaries replicate the primary's oplog and can become primary after an election.
- oplog
- Related: replica set, majority commit point, initial sync
A capped collection in the local database that records all data-modifying operations. Secondaries tail it to replicate changes and to catch up after network partitions or restarts.
- election
- Related: primary, priority, arbiter
The process by which replica set members vote to choose a new primary when the current primary is unavailable. Requires a majority of voting members and uses a term number to order attempts.
- arbiter
- Related: election, replica set
A replica set member that participates in elections but holds no data and cannot become primary. Used to provide a voting majority without the cost of an extra data-bearing node.
- priority
- Related: election, hidden member
A replica set configuration value that influences which secondary is most likely to be elected primary. Higher values win; a priority of 0 marks a member as unable to become primary.
- write concern
- Related: majority commit point, journaling
The level of acknowledgment requested from MongoDB for a write operation. Options include w:1, w:majority, and w:0 (fire-and-forget), balancing durability against latency.
- read preference
- Related: write concern, replica set
A client setting that routes read operations to specific replica set members, such as primary, primaryPreferred, secondary, or nearest. Affects consistency and read latency.
- initial sync
- Related: oplog, replica set
The process by which a new replica set member copies the full data set from an existing member and then applies oplog entries to catch up. It can be time-consuming and I/O intensive.
- hidden member
- Related: priority, delayed member
A replica set member that is invisible to client applications and cannot become primary. It can still vote in elections and is often used for dedicated backup or reporting workloads.
- delayed member
- Related: hidden member, oplog
A replica set member that applies oplog entries after a configured delay, maintaining a historical snapshot. Useful for recovering from accidental data deletion or corruption.
- sharded cluster
- Related: shard, mongos, config server
A MongoDB deployment architecture that distributes data across multiple shards to support horizontal scaling. It consists of shards, config servers, and mongos routers.
- shard key
- Related: hashed shard key, chunk, mongos
One or more indexed fields that determine how documents are distributed across shards in a sharded cluster. Choosing a good shard key is critical for even data distribution and query targeting.
- chunk
- Related: balancer, shard key, mongos
A contiguous range of shard key values stored on a single shard. The balancer migrates chunks between shards to keep data evenly distributed.
- balancer
- Related: chunk, shard, mongos
A background process in a sharded cluster that monitors chunk distribution and migrates chunks between shards to balance data. It can be stopped during maintenance windows.
- mongos
- Related: config server, shard, shard key
A routing service for sharded clusters. It receives client requests, determines which shard(s) hold the required data using the config servers' metadata, and aggregates results.
- config server
- Related: mongos, sharded cluster
A replica set that stores metadata about the sharded cluster, including chunk ranges and shard locations. In modern versions it must run as a replica set (CSRS).
- hashed shard key
- Related: shard key, chunk
A shard key that uses a hash of a field value to distribute documents evenly across shards. It provides good write distribution but poor range-query targeting.
- WiredTiger
- Related: journaling, checkpoint, compression
The default storage engine for MongoDB since version 3.2. It provides document-level concurrency control, compression, and checkpointing for durability.
- checkpoint
- Related: WiredTiger, journaling
A WiredTiger operation that flushes modified data from memory to disk, creating a consistent on-disk snapshot. Checkpoints occur at intervals and on clean shutdown.
- journaling
- Related: WiredTiger, write concern, checkpoint
A write-ahead logging mechanism that ensures durability by recording changes in a journal before applying them to data files. Enabled by default in WiredTiger.
- compression
- Related: WiredTiger
A WiredTiger feature that reduces storage footprint by compressing data and indexes. Supported algorithms include snappy, zlib, and zstd, configurable per collection or globally.
- SCRAM
- Related: x.509 authentication, role-based access control
Salted Challenge Response Authentication Mechanism, the default authentication method in MongoDB. It verifies user identity using a username and password without transmitting the password in clear text.
- x.509 authentication
- Related: SCRAM, TLS
A MongoDB authentication method that uses TLS client certificates to verify identity. Often used for internal member authentication in replica sets and sharded clusters.
- role-based access control
- Related: SCRAM, privilege
MongoDB's authorization model where privileges are grouped into roles, and roles are assigned to users. Built-in roles cover common needs; custom roles can be defined.
- TLS
- Related: x.509 authentication, encryption at rest
Transport Layer Security, used to encrypt network traffic between MongoDB clients and servers, and between cluster members. Requires configuring certificates and CA files.
- encryption at rest
- Related: TLS, WiredTiger
Protection of stored data by encrypting database files. MongoDB Enterprise offers native encryption at rest using WiredTiger and a key management service.
- mongodump
- Related: mongorestore, oplog
A command-line utility that creates a binary export of a MongoDB database's contents. It is often used for backups and can target a specific database or collection.
- mongorestore
- Related: mongodump, point-in-time recovery
A command-line utility that loads data from a binary dump created by mongodump into a MongoDB instance. It can replay oplog entries for point-in-time recovery.
- point-in-time recovery
- Related: oplog, mongorestore
The process of restoring a backup and then replaying oplog entries up to a specific moment, allowing recovery from accidental data loss or corruption.
- mongosh
- Related: mongodump, mongorestore
The modern MongoDB Shell, a JavaScript-based interactive interface for administering and querying MongoDB deployments. It replaces the legacy mongo shell.
- Ops Manager
- Related: Cloud Manager, automation agent
MongoDB's on-premises management platform for monitoring, backing up, and automating MongoDB deployments. It provides a web interface and agents for automation.
- Cloud Manager
- Related: Ops Manager
A hosted MongoDB management service for monitoring, backup, and automation. It is similar to Ops Manager but delivered as a cloud service.
- profiler
- Related: explain plan, index
A MongoDB feature that records slow or all operations to the system.profile collection. Used to diagnose performance issues and identify inefficient queries.
- explain plan
- Related: profiler, index
A diagnostic output that shows how MongoDB executes a query, including index usage, documents examined, and execution stages. Helps optimize query performance.
- index
- Related: explain plan, profiler
A data structure that improves the speed of read operations by allowing MongoDB to locate documents without scanning the entire collection. Common types include single-field, compound, and multikey.
- rolling upgrade
- Related: replica set, election
A maintenance procedure that upgrades replica set members one at a time, maintaining availability. It typically involves upgrading secondaries before the primary.
- fcv
- Related: rolling upgrade
Feature Compatibility Version, a setting that controls which features are available and ensures compatibility during upgrades. It must be set appropriately before and after version upgrades.