Index capacity and cost in DynamoDB - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

How does a Global Secondary Index (GSI) affect write capacity?

In DynamoDB, when you add a Global Secondary Index (GSI) to a table, how does it impact the write capacity units (WCUs) consumed?

A Writes consume WCUs for the base table, but GSIs consume read capacity units (RCUs) instead.

B Writes only consume WCUs for the base table; GSIs do not affect write capacity.

C Writes to the table consume WCUs for both the base table and the GSI, effectively doubling the write capacity usage.

D Writes consume WCUs only for the GSI, not the base table.

💡 Hint

Think about how data is duplicated in GSIs.

When you write to a table with a GSI, DynamoDB writes the data to both the base table and the GSI, so write capacity is consumed for both.

❓ query_result

intermediate

Calculate read capacity units for a query on a Local Secondary Index (LSI)

You have a DynamoDB table with 4 KB items. You query a Local Secondary Index (LSI) and retrieve 3 items. How many read capacity units (RCUs) are consumed if you use eventually consistent reads?

A 2 RCUs

B 6 RCUs

C 1 RCUs

D 12 RCUs

💡 Hint

Each 4 KB item read with eventually consistent reads consumes 0.5 RCU.

Each 4 KB item read with eventually consistent reads consumes 0.5 RCU. For 3 items, total is 3 * 0.5 = 1.5, rounded up to 2 RCUs.

📝 Syntax

advanced

Identify the correct syntax to create a GSI with provisioned capacity

Which of the following AWS CLI commands correctly creates a Global Secondary Index (GSI) with provisioned read and write capacity?

DynamoDB

aws dynamodb update-table --table-name Music --attribute-definitions AttributeName=Artist,AttributeType=S AttributeName=AlbumTitle,AttributeType=S --global-secondary-index-updates '[{"Create":{"IndexName":"ArtistAlbumIndex","KeySchema":[{"AttributeName":"Artist","KeyType":"HASH"},{"AttributeName":"AlbumTitle","KeyType":"RANGE"}],"Projection":{"ProjectionType":"ALL"},"ProvisionedThroughput":{"ReadCapacityUnits":5,"WriteCapacityUnits":5}}}]'

A The command is missing the --provisioned-throughput flag outside the GSI definition.

B The command as shown is correct and will create the GSI with provisioned capacity.

C The KeySchema must have only one attribute for a GSI, so this command is invalid.

D The ProjectionType must be KEYS_ONLY for provisioned GSIs, so this command will fail.

💡 Hint

Check the JSON structure inside the --global-secondary-index-updates parameter.

The command correctly specifies the GSI creation with KeySchema, Projection, and ProvisionedThroughput inside the Create object.

❓ optimization

advanced

Optimize read capacity usage for a frequently queried attribute

You have a DynamoDB table with a large number of items. You frequently query by a non-key attribute. Which approach optimizes read capacity usage?

A Add the attribute as a sort key in the base table and query with a filter expression.

B Scan the entire table and filter results on the client side.

C Use a Local Secondary Index (LSI) on that attribute and query it.

D Create a Global Secondary Index (GSI) on that attribute and query the GSI instead of the base table.

💡 Hint

Think about how indexes reduce the amount of data scanned.

Creating a GSI on the attribute allows efficient queries using indexed keys, reducing read capacity usage compared to scanning.

🔧 Debug

expert

Diagnose unexpected high write capacity consumption with GSIs

A DynamoDB table has two GSIs. Suddenly, write capacity usage spikes unexpectedly. Which is the most likely cause?

A The GSIs have large projected attributes causing more data to be written and higher WCUs consumed.

B The table's read capacity units are too low, causing throttling on writes.

C The table's partition key was changed, invalidating the GSIs and causing retries.

D The GSIs are using eventually consistent reads, which double the write capacity usage.

💡 Hint

Consider how projected attributes affect write costs.

Large projected attributes in GSIs increase the amount of data written to the index, raising write capacity consumption.