Implementing Data Modeling Patterns

1. Using Attribute Pattern

Aspect	Detail
Use	Many similar fields differing per row
Model	Array of {key, value} pairs
Benefit	Index one path instead of many columns
MongoDB example	attrs: [{k:"color",v:"red"},{k:"size","M"}]

2. Using Extended Reference Pattern

Aspect	Detail
Use	Embed frequently-accessed fields from referenced entity
Trade-off	Read perf vs sync complexity
Example	order.customer_name + customer_id

3. Using Subset Pattern

Aspect	Detail
Use	Document grows unbounded (large arrays)
Solution	Keep recent N inline; rest in separate collection
Benefit	Limit doc size, faster reads of common slice
Example	movie.recent_reviews (10) + reviews collection

4. Using Computed Pattern

Aspect	Detail
Use	Read-heavy derived value (e.g., averages, totals)
Solution	Pre-compute & store; refresh on write or periodically
Trade-off	Read perf vs write cost / staleness

5. Using Bucket Pattern

Aspect	Detail
Use	High-frequency time-series writes
Solution	Group N points per parent doc (bucket)
Benefit	Fewer documents, better compression
Example	IoT readings — 1 doc per hour with array of samples

6. Using Outlier Pattern

Aspect	Detail
Use	Few documents dwarf the rest
Solution	Flag outlier, move overflow data to separate doc
Example	Most books have < 100 reviews; bestsellers have 100k

7. Using Schema Versioning Pattern

Aspect	Detail
Use	Schema-less DB with evolving structure
Solution	Embed schema_version field
Migration	Lazy: upgrade on read; or batch background

8. Using Polymorphic Pattern

Aspect	Detail
Use	Similar entities with varying fields in one collection
Solution	type field discriminator + type-specific fields
Example	vehicles: cars + trucks + motorcycles with shared base

9. Using Approximation Pattern

Aspect	Detail
Use	Exact count not needed (page views, likes)
Solution	Increment counter only every Nth event
Benefit	Reduces write load N×
Alternatives	HyperLogLog for cardinality, Bloom filter for membership

10. Using Tree Pattern

Model	Use
Parent reference	Simplest; recursive lookup
Child references (array)	Fast children query
Array of ancestors	Fast ancestor query
Materialized path	String-based path traversal
Nested set	Read-heavy, complex trees