Franck PachotDatabase Developer Advocate
All minibooks
Database field guide · 33
BSON emulation, query planning, and index execution
Understand DocumentDB implementations through their compatibility boundary, BSON representation, multikey rules, query planner, and PostgreSQL extension architecture.
Build the mental model before choosing the mechanism.
This minibook was AI-generated from Franck Pachot's archived blog posts. Links to the original articles are included for source context and verification.01
DocumentDB names cover systems with different storage engines and execution layers. Wire compatibility and accepted syntax do not guarantee identical null, array, transaction, index, or planner behavior.
02
A document API implemented over PostgreSQL parses document commands, represents BSON, translates operators, and maps them to PostgreSQL access methods. Each boundary can affect semantics and cost.
03
Document fields can be indexed through generalized access methods such as RUM, but equality, sort, and range work best when the physical key order matches the query shape.
04
An array can produce several index keys for one document. This changes uniqueness, compound-key combinations, covering eligibility, and whether predicates refer to the same array element.
05
A covering or distinct scan can answer selected grouping shapes from ordered index entries without reading every document. Eligibility depends on group key, accumulator, ordering, and projection.
06
A new query planner may alter index choice, pushdown, and explain vocabulary. Performance conclusions must therefore retain engine version, statistics state, query shape, and returned cardinality.
07
Concrete mechanics, diagnostic evidence, and executable patterns to carry into a real system.
Capture server identity, compatibility version, query, index definitions, and explain output before attributing behavior to DocumentDB.
db.runCommand({ buildInfo: 1 })
db.collection.getIndexes()
db.collection.find(query).explain('executionStats')Array elements can create multiple keys per document. Compare keys examined, documents examined, and returned rows for the exact compound predicate.
db.orders.find({lines: {$elemMatch: {sku: 'A', qty: {$gte: 2}}}})
.explain('executionStats')Place equality, sort, then range requirements into a deliberate ordered index and verify whether the implementation can preserve order without an extra sort.
db.events.createIndex({tenantId: 1, createdAt: -1, score: 1})
db.events.find({tenantId: 7, score: {$gte: 10}})
.sort({createdAt: -1}).limit(20).explain('executionStats')A distinct or covering group optimization is shape-specific. Compare examined keys and documents with and without the candidate index.
db.sales.aggregate([
{$sort: {category: 1, createdAt: -1}},
{$group: {_id: '$category', latest: {$first: '$createdAt'}}}
], {explain: true})08
Optional deep dives with the complete experiments and product-version context behind this guide.