MongoDB Data Modeler Associate key terms

Storing related data as nested subdocuments or arrays inside a single parent document so one read returns the whole aggregate, trading duplication and document growth for fewer queries.

Storing a link such as an _id value in one document that points to a separate document, keeping data normalized but requiring additional queries or $lookup to join.

Organizing data so each fact is stored once in its own collection, eliminating duplication at the cost of extra reads or aggregation stages to reassemble related records.

Deliberately duplicating data across documents or embedding it to match read patterns, accepting update complexity and possible inconsistency in exchange for faster, simpler queries.

A relationship where each record on one side maps to exactly one record on the other, commonly modeled by embedding the dependent fields or by a unique reference field.

A relationship where one parent record relates to multiple children; modeled by an array of embedded subdocuments when bounded, or by child documents holding the parent's _id when unbounded.

A relationship where records on both sides relate to multiple records on the other; modeled with arrays of references on one or both sides, or with a separate join collection.

An array field whose element count can grow without a practical limit, risking the 16 MB document size cap and poor index behavior; usually refactored into a separate collection.

Increase in a document's stored size as embedded arrays or subdocuments are added, which can force the server to relocate the document and degrade write performance.

The basic unit of storage in MongoDB, a BSON-encoded set of field-and-value pairs, including nested documents and arrays, identified by a unique _id field.

A named grouping of documents within a database that shares no enforced schema by default, allowing heterogeneous document shapes unless validation rules are applied.

A document nested as the value of a field inside another document, useful for grouping attributes that are always read and written together with the parent.

The binary serialization format MongoDB uses to store documents, supporting types beyond JSON such as ObjectId, Date, Decimal128, Binary, and 64-bit integers.

A 12-byte BSON type used as the default _id value, combining a timestamp, random per-process value, and counter, so generated identifiers are roughly time-ordered.

The mandatory unique identifier of every document; its value must be unique within the collection and an index on it is created automatically.

A named key in a document paired with a value; field names are strings, and the same name may hold different BSON types across documents in schemaless collections.

Path syntax using dots to address nested fields and array elements, as in address.city or items.0.sku, used in queries, updates, projections, and index keys.

Collection-level rules using $jsonSchema or query operators that reject inserts and updates not matching declared field types, required fields, or allowed value ranges.

A design where documents in one collection share some common fields but differ in shape or subtype, often distinguished by a discriminator field to support varied entities.

A reusable modeling technique, such as subset, bucket, or computed, applied to match document structure to known query and write access patterns.

The set of queries, updates, and sorts an application performs most often; identifying these first is the central step in MongoDB schema design.

An approach that shapes collections, embedded fields, and indexes around how data is read and written rather than around normalized entity relationships alone.

Keeping a small, frequently accessed subset of a large related dataset embedded in the parent document while the full set lives in another collection, reducing working set size.

Grouping many small time-series or event records into a bounded number of documents, each holding a batch of entries, to reduce document count and index size.

Precalculating values such as totals, counts, or averages and storing them in the document so reads avoid expensive aggregation, at the cost of keeping them updated on writes.

Embedding a few frequently needed fields copied from a referenced document, avoiding a join for common reads while the authoritative copy stays in its own collection.

Moving a large, sparse set of similar fields into an array of key-value subdocuments so each attribute can be indexed uniformly and queried by name.

The portion of data and indexes actively used by queries and writes at a given time; keeping it within RAM is a primary goal of schema and index design.

The number of distinct values a field holds across documents; high-cardinality fields generally make more selective index keys than low-cardinality ones.

How well an index key narrows the candidate documents for a query; highly selective indexes filter out most documents and reduce the work of a scan.

An index on one field, in ascending or descending order, that speeds equality, range, and sort operations using that field and can be multikey if the field is an array.

An index on multiple fields in a specified order; it supports queries on a prefix of its keys and follows the ESR guideline for equality, sort, and range fields.

A rule for ordering compound index keys: equality fields first, then fields used for sort, then range fields, to maximize how much of the index a query can use efficiently.

The leading field or fields of a compound index; a query can use the index only if its equality and sort conditions align with a usable prefix.

An index created on a field containing an array, with one index entry per array element, enabling efficient queries on array membership and element values.

An index that rejects inserts or updates producing duplicate key values, commonly used on natural keys and on the _id field, which is unique by default.

An index that includes only documents matching a filter expression, reducing index size and write cost when queries target a well-defined subset of the collection.

An index that omits entries for documents missing the indexed field, saving space but only usable by queries that exclude documents lacking that field.

A special single-field index on a Date value that lets the server automatically delete documents after a specified number of seconds, useful for expiring sessions or logs.

A query fully satisfied from index keys alone, with no document fetch, possible when all queried and returned fields are part of the index and no field is excluded.

Specifying which fields a query returns, reducing network and memory use and, when only indexed fields are requested, enabling a covered query.

A collection type that stores measurements with a time field and metadata field, automatically organizing data into buckets and supporting optimized storage and queries.