DynamoDB Streams concept

Introduction

DynamoDB Streams lets you see changes made to your database in real time. It helps you react to data updates quickly and keep things in sync.

You want to track every change made to your database items.

You need to update other systems automatically when your data changes.

You want to keep a backup or audit trail of all changes.

You want to trigger notifications or workflows when data updates happen.

You want to replicate data to another database or region.

Syntax

DynamoDB

Enable DynamoDB Streams on your table with a stream view type:

- KEYS_ONLY
- NEW_IMAGE
- OLD_IMAGE
- NEW_AND_OLD_IMAGES

The stream view type controls what data about the change is recorded.

Streams capture inserts, updates, and deletes on your table.

Examples

This records only the primary key attributes of changed items.

DynamoDB

Enable stream with KEYS_ONLY view type

This records the entire item after the change.

DynamoDB

Enable stream with NEW_IMAGE view type

This records the entire item before the change.

DynamoDB

Enable stream with OLD_IMAGE view type

This records both the item before and after the change.

DynamoDB

Enable stream with NEW_AND_OLD_IMAGES view type

Sample Program

This command enables DynamoDB Streams on the 'MusicCollection' table. It records both the old and new versions of items when they change.

DynamoDB

aws dynamodb update-table \
  --table-name MusicCollection \
  --stream-specification StreamEnabled=true,StreamViewType=NEW_AND_OLD_IMAGES

Important Notes

Enabling streams may slightly increase your costs because of extra storage and processing.

You can use AWS Lambda to automatically process stream records as they arrive.

Stream records are kept for 24 hours before they expire.

Summary

DynamoDB Streams capture changes to your table in real time.

You choose what data to capture using stream view types.

Streams help you build reactive and synchronized applications.

Practice

1. What is the primary purpose of DynamoDB Streams?

easy

A. To back up the entire DynamoDB table automatically every hour

B. To capture and record changes made to items in a DynamoDB table in real time

C. To increase the read capacity of a DynamoDB table

D. To encrypt data stored in DynamoDB tables

Solution

1. Step 1: Understand DynamoDB Streams functionality

   DynamoDB Streams capture data changes such as inserts, updates, and deletes on a table.

2. Step 2: Compare options with Streams purpose

   Only To capture and record changes made to items in a DynamoDB table in real time describes capturing changes in real time, which is the core function of Streams.

3. Final Answer:

   To capture and record changes made to items in a DynamoDB table in real time -> Option B

4. Quick Check:

   DynamoDB Streams = capture changes [OK]

Hint: Streams track table changes live, not backups or capacity [OK]

Common Mistakes:

• Confusing Streams with backups
• Thinking Streams increase capacity
• Assuming Streams encrypt data

2. Which of the following is the correct way to enable DynamoDB Streams on a table?

easy

A. Enable Auto Scaling on the table

B. Add a Global Secondary Index with StreamEnabled = true

C. Create a Lambda function and attach it to the table without enabling Streams

D. Set the StreamSpecification with StreamEnabled = true and choose a StreamViewType

Solution

1. Step 1: Recall how to enable Streams

   Streams are enabled by setting StreamSpecification with StreamEnabled true and selecting a StreamViewType.

2. Step 2: Evaluate options

   Set the StreamSpecification with StreamEnabled = true and choose a StreamViewType correctly describes enabling Streams. Other options relate to indexes, Lambda, or scaling, which do not enable Streams.

3. Final Answer:

   Set the StreamSpecification with StreamEnabled = true and choose a StreamViewType -> Option D

4. Quick Check:

   Enable Streams = StreamSpecification true + StreamViewType [OK]

Hint: Enable Streams by setting StreamSpecification true with view type [OK]

Common Mistakes:

• Confusing Streams with indexes
• Assuming Lambda enables Streams automatically
• Thinking Auto Scaling enables Streams

3. Given a DynamoDB Stream with StreamViewType set to NEW_AND_OLD_IMAGES, what data will each stream record contain?

medium

A. Only the new image of the item after the change

B. Only the old image of the item before the change

C. Both the new and old images of the item

D. Only the keys of the changed item

Solution

1. Step 1: Understand StreamViewType NEW_AND_OLD_IMAGES

   This view type captures both the item image before and after the change.

2. Step 2: Match options to the view type

   Both the new and old images of the item correctly states both new and old images are included. Other options only mention one or keys.

3. Final Answer:

   Both the new and old images of the item -> Option C

4. Quick Check:

   NEW_AND_OLD_IMAGES = new + old images [OK]

Hint: NEW_AND_OLD_IMAGES means both before and after data included [OK]

Common Mistakes:

• Thinking only new or only old image is included
• Confusing keys-only view with full images
• Ignoring the StreamViewType setting

4. You have enabled DynamoDB Streams but your Lambda function is not receiving any events despite recent writes to the table. What is a possible reason?

medium

A. The Lambda function is not properly configured as an event source for the stream

B. The table has no write activity, so no stream records are generated

C. All of the above

D. The StreamViewType was set to KEYS_ONLY, but your Lambda expects full images

Solution

1. Step 1: Analyze the scenario

   Streams enabled with recent writes means records are generated, so issue is likely with Lambda setup to consume stream events.

2. Step 2: Evaluate options

   The Lambda not being configured as an event source for the stream explains no events. All of the above doesn't apply since writes occurred. No write activity contradicted by scenario. StreamViewType affects record content, not event delivery to Lambda.

3. Final Answer:

   The Lambda function is not properly configured as an event source for the stream -> Option A

4. Quick Check:

   No events despite writes = Lambda event source not mapped [OK]

Hint: Lambda must be mapped as event source to DynamoDB Stream [OK]

Common Mistakes:

• Ignoring that no writes means no stream records
• Forgetting to attach Lambda as event source
• Assuming KEYS_ONLY always sends full data

5. You want to build a system that reacts only when an item's attribute status changes in your DynamoDB table. Which StreamViewType should you choose to efficiently detect this change?

hard

A. NEW_AND_OLD_IMAGES

B. OLD_IMAGE

C. NEW_IMAGE

D. KEYS_ONLY

Solution

1. Step 1: Identify data needed to detect attribute changes

   To know if 'status' changed, you need both old and new values to compare.

2. Step 2: Match StreamViewType to requirement

   Only NEW_AND_OLD_IMAGES provides both before and after images for comparison.

3. Final Answer:

   NEW_AND_OLD_IMAGES -> Option A

4. Quick Check:

   Detect attribute change = need old + new images [OK]

Hint: Compare old and new images to detect attribute changes [OK]

Common Mistakes:

• Choosing only NEW_IMAGE or OLD_IMAGE which lacks comparison data
• Assuming KEYS_ONLY includes attribute data
• Not realizing both images are needed to detect changes