DEVPREP CODEX · #12

MongoDB Interview Questions

171 curated questions graded from Foundations (Easy) to Practical Patterns (Medium) and Internals & Architecture (Hard).

TOTAL QUESTIONS

171

THEORY QUESTIONS

150

IMPLEMENTATION FOLIOS

FREE QUESTIONS

5 / Level

Easy Level·57 Questions Total

Foundations & Core Concepts

Open Full Easy List

Q1What is MongoDB and how does it differ from traditional RDBMS?

Definition: MongoDB is a document-oriented NoSQL database that stores data in flexible, JSON-like documents (BSON format).
Schema Difference: Traditional RDBMS uses a rigid schema with tables, rows, and columns. MongoDB uses a schema-less structure with collections and documents, allowing varying structures within the same collection.
Relationships: RDBMS relies heavily on normalized tables and relations via foreign keys and joins. MongoDB relies on denormalized embedded documents or manual references, optimizing for read-heavy operations.
Scalability: RDBMS typically scales vertically (adding hardware power), while MongoDB scales horizontally (sharding across multiple servers).

Q2What is BSON and how is it different from JSON?

Definition: BSON stands for Binary JSON. It is a binary-serialized representation of JSON-like documents.
Size & Efficiency: BSON is designed for high speed and space efficiency. It is faster to parse and traverse compared to JSON.
Data Types: BSON extends JSON by supporting additional data types such as Date, ObjectId, BinData (binary data), Double, and 32-bit/64-bit integers. JSON only natively supports strings, numbers, booleans, null, arrays, and objects.

Q3Explain the structure of a MongoDB ObjectId.

Definition: A unique 12-byte identifier automatically generated as the primary key (_id) for each document.
Byte Breakdown:
- First 4 bytes: Unix timestamp (seconds since epoch) representing the creation time.
- Next 5 bytes: Random value unique to the machine and process (ensures uniqueness across distributed systems).
- Last 3 bytes: An incrementing counter, initialized with a random value.
Implication: ObjectIds are naturally ordered by creation time, allowing sorting by _id to act as a chronological sort.

Q4What is a dynamic schema and what are its advantages?

Concept: MongoDB collections do not enforce document structure. Documents in the same collection can have different fields, types, and nested nesting.
Advantages:
- Agility: Rapid development cycles since database migrations or schema alterations (ALTER TABLE) are not required.
- Polymorphism: Ability to represent diverse entities or semi-structured data without complex join tables.
- No Null Fields: Eliminates the need to store null values for optional attributes, saving storage space.

Q5What are collections and documents in MongoDB?

Collection: Equivalent to an RDBMS table. It is a grouping of MongoDB documents. Collections do not enforce schemas but can have schema validation rules if configured.
Document: Equivalent to an RDBMS row. It is the basic unit of data in MongoDB, composed of key-value pairs stored in BSON format.

52 More Easy Questions Locked

Unlock the complete MongoDB Easy question bank

Get instant access to all 57 questions, in-depth model answers, code sandboxes, and all 27+ technologies for a single one-time payment.

Unlock All — One-time payment · Lifetime access

Medium Level·57 Questions Total

Mechanisms & Practical Patterns

Open Full Medium List

Q1Explain the components of MongoDB aggregation framework ($match, $project, $group, $unwind, $lookup, $out).

$match: Filters documents to pass only matching documents to the next stage (equivalent to SQL WHERE). Must be placed as early as possible to utilize indexes.
$project: Reshapes documents by adding, renaming, or removing fields (equivalent to SQL SELECT).
$group: Groups documents by a specified key and performs accumulative operations (like $sum, $avg, $min, $max) on the grouped data.
$unwind: Deconstructs an array field from input documents to output a document for each element of the array.
$lookup: Performs a left outer join to another collection in the same database to filter/add data.
$out: Writes the resulting documents of the aggregation pipeline to a new or existing collection (must be the final stage).

Q2What is the difference between `$merge` and `$out` stages in aggregation?

$out:
- Overwrites the entire output collection if it exists, or creates a new one.
- Cannot write to a sharded collection.
- Replaces the collection atomically upon pipeline completion.
$merge:
- Introduces on-demand incremental updates. It can merge results (insert, update, discard, or fail on matches) into an existing target collection.
- Can write to sharded or unsharded collections, even across different databases.

Q3How does `$lookup` perform a join? What are its performance characteristics and index requirements?

Mechanism: Acts as a nested loop join. For each document in the source collection, it queries the target collection.
Performance: Can be extremely slow if the target collection is large.
Index Requirements: The joined foreign field on the target collection must be indexed. If not, MongoDB must perform a full collection scan on the target collection for every single document processed by $lookup.

Q4What is the difference between single-field indexing and compound indexing regarding query coverage?

Single-Field Index: Indexes a single field (e.g., { a: 1 }). Optimizes queries filtering or sorting on a.
Compound Index: Indexes multiple fields (e.g., { a: 1, b: -1 }). Optimizes queries filtering on a, or a and b.
Index Prefix Rule: A compound index { a, b } covers queries on a but not queries filtering only on b. To optimize queries on b, a separate single-field or compound index starting with b is required.

Q5Explain the performance implications of indexing array fields (multikey indexes).

Multikey Index: Automatically created when an index is created on a field containing an array.
Under the Hood: MongoDB creates separate index keys for every single element in the array.
Implications:
- High write overhead: If a document has an array with 100 elements, inserting it creates 100 index entries.
- Storage bloat: Multikey indexes occupy significantly more RAM and disk storage.
- Limit: A compound multikey index cannot have more than one array field indexed in the index definition to prevent exponential explosion of index entries.

52 More Medium Questions Locked

Unlock the complete MongoDB Medium question bank

Get instant access to all 57 questions, in-depth model answers, code sandboxes, and all 27+ technologies for a single one-time payment.

Unlock All — One-time payment · Lifetime access

Hard Level·57 Questions Total

Internals & System Architecture

Open Full Hard List

Q1Explain WiredTiger storage engine internals, specifically MVCC, page structure, checkpointing, and reconciliation.

Multi-Version Concurrency Control (MVCC): WiredTiger handles concurrent transactions by keeping multiple versions of modified documents. Readers get a point-in-time snapshot, avoiding read locks while writes occur concurrently.
Page Structure: Data is organized in B-Tree pages (internal and leaf pages) in memory and on disk. Modifications are stored as in-memory "update lists" appended to the original on-disk page layout.
Checkpointing: Every 60 seconds (or when journal writes reach 2GB), WiredTiger flushes all dirty in-memory pages to disk, creating a crash-consistent checkpoint.
Reconciliation: During checkpointing, the engine combines the in-memory update lists with the old on-disk page format, compiles them into a new layout, compresses it, and writes clean, new pages to disk.

Q2What is read concern `snapshot` and how does it guarantee read isolation in multi-document transactions?

Concept: Only available within multi-document sessions/transactions.
Guarantee: It provides strict snapshot isolation. The query engine reads data from a synchronized, point-in-time snapshot of the database across all replica set nodes.
Mechanism: Ensures that all operations inside the transaction observe a single, consistent state of the database. Even if other clients commit transactions concurrently, the transaction's view remains completely isolated and unchanged.

Q3How does the election protocol version 1 (pv1) prevent split-brain scenarios in a distributed replica set?

Split-Brain: A situation where network partitions cause two sub-groups of a replica set to both think they have an active primary, leading to divergent write logs.
PV1 Solution: Uses strict raft-like consensus. A candidate node must receive a majority vote of all voting nodes (e.g., at least 2 out of 3, or 3 out of 5) to become primary.
Double Voting Prevention: A voter can only vote once per election term (using term increment counters). If a network partition occurs, only the partition containing a true majority can elect a primary; the minority partition is physically blocked from electing a primary.

Q4Explain the step-down process of a Primary node and how to configure priority and votes parameters.

Step-Down: An administrative command (replSetStepDown) or automatic behavior where the active primary relinquishes its role and becomes a secondary.
Trigger: Happens during maintenance, or when the primary detects that it has lost connectivity to a majority of the replica set nodes.
Configuration:
- priority: A number (0 to 1000) indicating preference for primary. Secondary nodes with priority: 0 can never become primary.
- votes: A boolean (0 or 1) indicating voting rights. Up to 7 members in a replica set can have votes.

Q5How does MongoDB handle node synchronization from scratch (Initial Sync)?

Phase 1: Metadata & Cloning: The syncing node creates database structures, drops existing databases, clones all collections and indexes from a sync source node, while concurrently logging any new oplog writes occurring on the sync source.
Phase 2: Oplog Application: The syncing node applies the recorded oplog writes to bring its local dataset up to speed.
Pitfalls: If the sync source's oplog rolls over (overwrites older records) before Phase 1 completes, the sync fails, requiring a complete restart.

52 More Hard Questions Locked

Unlock the complete MongoDB Hard question bank

Get instant access to all 57 questions, in-depth model answers, code sandboxes, and all 27+ technologies for a single one-time payment.

Unlock All — One-time payment · Lifetime access

DEDICATED LEVEL PAGES

Browse All MongoDB Questions by Difficulty

Easy57 Questions

Core concepts, definitions, basic syntax, and first principles expected in round 1 screening.

VIEW FULL LIST

Medium57 Questions

Real-world mechanisms, state management, edge cases, performance trade-offs, and practical coding.

VIEW FULL LIST

Hard57 Questions

Deep runtime internals, memory models, distributed design, concurrency failure modes, and architectural decisions.

VIEW FULL LIST

INTERNAL LINKING & DEPENDENCIES

Related Technologies to MongoDB

CAREER PATHWAYS

MongoDB is a key requirement for these engineering roles:

Backend Engineer Guide Full Stack Engineer Guide

DEVPREP CODEX · #12

MongoDB Interview Questions

171 curated questions graded from Foundations (Easy) to Practical Patterns (Medium) and Internals & Architecture (Hard).

TOTAL QUESTIONS

171

THEORY QUESTIONS

150

IMPLEMENTATION FOLIOS

FREE QUESTIONS

5 / Level

Easy Level·57 Questions Total

Foundations & Core Concepts

Open Full Easy List

Q1What is MongoDB and how does it differ from traditional RDBMS?

Definition: MongoDB is a document-oriented NoSQL database that stores data in flexible, JSON-like documents (BSON format).
Schema Difference: Traditional RDBMS uses a rigid schema with tables, rows, and columns. MongoDB uses a schema-less structure with collections and documents, allowing varying structures within the same collection.
Relationships: RDBMS relies heavily on normalized tables and relations via foreign keys and joins. MongoDB relies on denormalized embedded documents or manual references, optimizing for read-heavy operations.
Scalability: RDBMS typically scales vertically (adding hardware power), while MongoDB scales horizontally (sharding across multiple servers).

Q2What is BSON and how is it different from JSON?

Definition: BSON stands for Binary JSON. It is a binary-serialized representation of JSON-like documents.
Size & Efficiency: BSON is designed for high speed and space efficiency. It is faster to parse and traverse compared to JSON.
Data Types: BSON extends JSON by supporting additional data types such as Date, ObjectId, BinData (binary data), Double, and 32-bit/64-bit integers. JSON only natively supports strings, numbers, booleans, null, arrays, and objects.

Q3Explain the structure of a MongoDB ObjectId.

Definition: A unique 12-byte identifier automatically generated as the primary key (_id) for each document.
Byte Breakdown:
- First 4 bytes: Unix timestamp (seconds since epoch) representing the creation time.
- Next 5 bytes: Random value unique to the machine and process (ensures uniqueness across distributed systems).
- Last 3 bytes: An incrementing counter, initialized with a random value.
Implication: ObjectIds are naturally ordered by creation time, allowing sorting by _id to act as a chronological sort.

Q4What is a dynamic schema and what are its advantages?

Concept: MongoDB collections do not enforce document structure. Documents in the same collection can have different fields, types, and nested nesting.
Advantages:
- Agility: Rapid development cycles since database migrations or schema alterations (ALTER TABLE) are not required.
- Polymorphism: Ability to represent diverse entities or semi-structured data without complex join tables.
- No Null Fields: Eliminates the need to store null values for optional attributes, saving storage space.

Q5What are collections and documents in MongoDB?

Collection: Equivalent to an RDBMS table. It is a grouping of MongoDB documents. Collections do not enforce schemas but can have schema validation rules if configured.
Document: Equivalent to an RDBMS row. It is the basic unit of data in MongoDB, composed of key-value pairs stored in BSON format.

52 More Easy Questions Locked

Unlock the complete MongoDB Easy question bank

Get instant access to all 57 questions, in-depth model answers, code sandboxes, and all 27+ technologies for a single one-time payment.

Unlock All — One-time payment · Lifetime access

Medium Level·57 Questions Total

Mechanisms & Practical Patterns

Open Full Medium List

Q1Explain the components of MongoDB aggregation framework ($match, $project, $group, $unwind, $lookup, $out).

$match: Filters documents to pass only matching documents to the next stage (equivalent to SQL WHERE). Must be placed as early as possible to utilize indexes.
$project: Reshapes documents by adding, renaming, or removing fields (equivalent to SQL SELECT).
$group: Groups documents by a specified key and performs accumulative operations (like $sum, $avg, $min, $max) on the grouped data.
$unwind: Deconstructs an array field from input documents to output a document for each element of the array.
$lookup: Performs a left outer join to another collection in the same database to filter/add data.
$out: Writes the resulting documents of the aggregation pipeline to a new or existing collection (must be the final stage).

Q2What is the difference between `$merge` and `$out` stages in aggregation?

$out:
- Overwrites the entire output collection if it exists, or creates a new one.
- Cannot write to a sharded collection.
- Replaces the collection atomically upon pipeline completion.
$merge:
- Introduces on-demand incremental updates. It can merge results (insert, update, discard, or fail on matches) into an existing target collection.
- Can write to sharded or unsharded collections, even across different databases.

Q3How does `$lookup` perform a join? What are its performance characteristics and index requirements?

Mechanism: Acts as a nested loop join. For each document in the source collection, it queries the target collection.
Performance: Can be extremely slow if the target collection is large.
Index Requirements: The joined foreign field on the target collection must be indexed. If not, MongoDB must perform a full collection scan on the target collection for every single document processed by $lookup.

Q4What is the difference between single-field indexing and compound indexing regarding query coverage?

Single-Field Index: Indexes a single field (e.g., { a: 1 }). Optimizes queries filtering or sorting on a.
Compound Index: Indexes multiple fields (e.g., { a: 1, b: -1 }). Optimizes queries filtering on a, or a and b.
Index Prefix Rule: A compound index { a, b } covers queries on a but not queries filtering only on b. To optimize queries on b, a separate single-field or compound index starting with b is required.

Q5Explain the performance implications of indexing array fields (multikey indexes).

Multikey Index: Automatically created when an index is created on a field containing an array.
Under the Hood: MongoDB creates separate index keys for every single element in the array.
Implications:
- High write overhead: If a document has an array with 100 elements, inserting it creates 100 index entries.
- Storage bloat: Multikey indexes occupy significantly more RAM and disk storage.
- Limit: A compound multikey index cannot have more than one array field indexed in the index definition to prevent exponential explosion of index entries.

52 More Medium Questions Locked

Unlock the complete MongoDB Medium question bank

Get instant access to all 57 questions, in-depth model answers, code sandboxes, and all 27+ technologies for a single one-time payment.

Unlock All — One-time payment · Lifetime access

Hard Level·57 Questions Total

Internals & System Architecture

Open Full Hard List

Q1Explain WiredTiger storage engine internals, specifically MVCC, page structure, checkpointing, and reconciliation.

Multi-Version Concurrency Control (MVCC): WiredTiger handles concurrent transactions by keeping multiple versions of modified documents. Readers get a point-in-time snapshot, avoiding read locks while writes occur concurrently.
Page Structure: Data is organized in B-Tree pages (internal and leaf pages) in memory and on disk. Modifications are stored as in-memory "update lists" appended to the original on-disk page layout.
Checkpointing: Every 60 seconds (or when journal writes reach 2GB), WiredTiger flushes all dirty in-memory pages to disk, creating a crash-consistent checkpoint.
Reconciliation: During checkpointing, the engine combines the in-memory update lists with the old on-disk page format, compiles them into a new layout, compresses it, and writes clean, new pages to disk.

Q2What is read concern `snapshot` and how does it guarantee read isolation in multi-document transactions?

Concept: Only available within multi-document sessions/transactions.
Guarantee: It provides strict snapshot isolation. The query engine reads data from a synchronized, point-in-time snapshot of the database across all replica set nodes.
Mechanism: Ensures that all operations inside the transaction observe a single, consistent state of the database. Even if other clients commit transactions concurrently, the transaction's view remains completely isolated and unchanged.

Q3How does the election protocol version 1 (pv1) prevent split-brain scenarios in a distributed replica set?

Split-Brain: A situation where network partitions cause two sub-groups of a replica set to both think they have an active primary, leading to divergent write logs.
PV1 Solution: Uses strict raft-like consensus. A candidate node must receive a majority vote of all voting nodes (e.g., at least 2 out of 3, or 3 out of 5) to become primary.
Double Voting Prevention: A voter can only vote once per election term (using term increment counters). If a network partition occurs, only the partition containing a true majority can elect a primary; the minority partition is physically blocked from electing a primary.

Q4Explain the step-down process of a Primary node and how to configure priority and votes parameters.

Step-Down: An administrative command (replSetStepDown) or automatic behavior where the active primary relinquishes its role and becomes a secondary.
Trigger: Happens during maintenance, or when the primary detects that it has lost connectivity to a majority of the replica set nodes.
Configuration:
- priority: A number (0 to 1000) indicating preference for primary. Secondary nodes with priority: 0 can never become primary.
- votes: A boolean (0 or 1) indicating voting rights. Up to 7 members in a replica set can have votes.

Q5How does MongoDB handle node synchronization from scratch (Initial Sync)?

Phase 1: Metadata & Cloning: The syncing node creates database structures, drops existing databases, clones all collections and indexes from a sync source node, while concurrently logging any new oplog writes occurring on the sync source.
Phase 2: Oplog Application: The syncing node applies the recorded oplog writes to bring its local dataset up to speed.
Pitfalls: If the sync source's oplog rolls over (overwrites older records) before Phase 1 completes, the sync fails, requiring a complete restart.

52 More Hard Questions Locked