DynamoDB crash course: part 3 - design patterns
Briefly

Composite (synthetic) keys enable access patterns by composing table and index keys from other attributes, typically via string concatenation with padding for sort keys. Denormalization trade-offs arise when embedding attributes in sort keys, requiring either immutable separate attributes for indexing or parsing embedded values to avoid inconsistency. DynamoDB added multi-attribute key support to GSIs in 2025, while tables did not receive the same capability. AWS guidance recommends maintaining as few tables as possible, noting that a single table with inverted indexes can support complex hierarchical queries. Single table design stores multiple entity types in one table and distinguishes them using key prefixes, with one table corresponding to an application’s data model.
Read at death and gravity
Unable to calculate read time
[
|
]