DynamoDB Streams

• Streams here are ordered list of item level modification i.e., (create/update/delete) in a table.

• Stream records can be sent to,

  • Kinesis Data Streams

  • AWS Lambda

  • KCL Applications

• Data Retention for upto 24 hours

• Use cases include,

  • React to changes in real-time.

  • Analytics

  • Insert into derivative tables.

  • Insert into OpenSearch service.

  • Implement cross-region replication.

• Provides ability to choose the information that will be written to streams. The information includes,

  Attribute Name	Description
  KEYS_ONLY	Only attributes that will be modified
  NEW_IMAGE	The entire items, as it appears after it was modified.
  OLD_IMAGE	The entire items, as it appears before it was modified.
  NEW_AND_OLD_IMAGES	Both new and old images of the item.

• Streams are made of shards, just like Kinesis Data Streams.

• You don't need to provisions shards, this is automated by AWS.

• Records are not retroactively populated in a stream after enabling it.

Integration with Lambda

• Define an Event Source Mapping to read from a DynamoDB Streams.

• You need to ensure the lambda function have appropriate function.

• Event Source Mapping will poll the data from the DynamoDB streams.

• Trigger need to be set up so that lambda will be triggered.

• Once above set up is done, Lambda will be ready to invoked synchronously for a batch of record fetched from DynamoDB streams.