DynamoDB Glossary

A DynamoDB design pattern where each node in a graph stores a reference to its adjacent nodes (parent, children, or related entities) using partition key and sort key combinations.

A composite key combines a partition key and a sort key into one primary key. The term also covers composite attributes, where several values are concatenated into a single key.

Using one Global Secondary Index to serve multiple entity types by writing different values to the same GSI key attributes depending on the entity type.

A hot partition is a single DynamoDB partition receiving disproportionate traffic, causing throttling even when the table has capacity to spare.

All items in a DynamoDB table (or local secondary index) that share the same partition key value. Item collections are bounded by a 10 GB per-partition limit.

Using generic DynamoDB attribute names (pk, sk, gsi1pk) to store different logical values for different entity types in a single-table design.

The partition key is the attribute DynamoDB hashes to decide where an item is stored. It determines data distribution and constrains every query you can run.

Single-table design stores multiple entity types in one DynamoDB table so related data can be fetched in a single query. Here's how it works and when it's worth it.

The sort key orders items within a partition and enables range queries. Its design determines which access patterns your schema can serve without a GSI.

A sparse index is a GSI that only contains items which have the index key attribute set. Use it to query a small subset of a large table cheaply.

Write sharding spreads traffic for one logical partition key across several physical partitions by appending a shard suffix, avoiding hot partition throttling.