Partition key selection in DynamoDB - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Why is choosing a good partition key important in DynamoDB?

Imagine you have a DynamoDB table storing user orders. Which reason best explains why selecting a good partition key matters?

A It prevents the need for a sort key in the table design.

B It guarantees that all data is stored in a single partition for faster access.

C It allows DynamoDB to automatically create indexes without user input.

D It ensures data is evenly distributed across partitions to avoid hot spots and improve performance.

💡 Hint

Think about how DynamoDB stores data internally and what happens if too many items share the same partition key.

A good partition key spreads data evenly across partitions, avoiding hot spots that slow down reads and writes. Options B, C, and D are incorrect because data is distributed, indexes require design, and sort keys are optional but unrelated to partition key choice.

❓ query_result

intermediate

What is the result of this partition key choice?

Given a DynamoDB table with partition key 'Country' and sort key 'UserID', what happens if most users are from the same country?

DynamoDB

Table: Users
Partition Key: Country
Sort Key: UserID
Data: 90% users have Country = 'USA', 10% others

Question: How does this affect performance?

A The 'USA' partition becomes a hot partition causing throttling and slower queries.

B Data is evenly distributed because sort key balances the load.

C DynamoDB automatically splits the 'USA' partition to handle load.

D Queries on 'USA' users will be faster due to data locality.

💡 Hint

Consider how DynamoDB partitions data based on the partition key only.

Since partition key is 'Country', all 'USA' users share the same partition, causing a hot partition. Sort key does not affect partition distribution. DynamoDB does not automatically split hot partitions instantly. Queries on hot partitions slow down.

📝 Syntax

advanced

Which partition key definition is valid in DynamoDB table creation?

Choose the correct AWS CLI command snippet to create a DynamoDB table with a partition key named 'UserID' of type string.

DynamoDB

aws dynamodb create-table --table-name Users --attribute-definitions ... --key-schema ... --provisioned-throughput ReadCapacityUnits=5,WriteCapacityUnits=5

A --attribute-definitions AttributeName=UserID,AttributeType=N --key-schema AttributeName=UserID,KeyType=HASH

B --attribute-definitions AttributeName=UserID,AttributeType=S --key-schema AttributeName=UserID,KeyType=HASH

C --attribute-definitions AttributeName=UserID,AttributeType=BOOL --key-schema AttributeName=UserID,KeyType=HASH

D --attribute-definitions AttributeName=UserID,AttributeType=SS --key-schema AttributeName=UserID,KeyType=HASH

💡 Hint

Check the valid attribute types for partition keys in DynamoDB.

Partition keys must be of type String (S), Number (N), or Binary (B). BOOL and SS (string set) are invalid for keys. Option B uses String type correctly.

❓ optimization

advanced

How to optimize partition key for a high-traffic IoT sensor data table?

You have a DynamoDB table storing sensor readings. Each sensor sends data every second. Which partition key design best avoids hot partitions?

A Use 'SensorID' as partition key to distribute data evenly across sensors.

B Use a constant value as partition key to simplify queries.

C Use 'SensorID' concatenated with date (e.g., 'SensorID#20240601') as partition key.

D Use 'Timestamp' as partition key to group data by time.

💡 Hint

Think about how to spread writes evenly when sensors send frequent data.

Using only 'SensorID' may cause hot partitions if some sensors send more data. Using timestamp alone groups all data by time causing hot spots. A constant partition key causes a single hot partition. Concatenating 'SensorID' with date creates more partitions and balances load.

🔧 Debug

expert

Why does this DynamoDB query cause throttling despite low overall traffic?

Given a table with partition key 'UserID' and sort key 'OrderDate', a query filters orders for UserID='123'. Despite low total traffic, the query is throttled. What is the likely cause?

DynamoDB

Query:
{
  TableName: 'Orders',
  KeyConditionExpression: 'UserID = :uid',
  ExpressionAttributeValues: { ':uid': '123' }
}

A UserID='123' has many orders causing a hot partition and throttling.

B The query syntax is invalid causing throttling errors.

C Sort key 'OrderDate' is missing in the query causing throttling.

D Provisioned throughput is too high causing throttling.

💡 Hint

Consider what happens if one partition key value has a lot of data.

Throttling happens because the partition for UserID='123' is overloaded with many orders. Query syntax is valid. High provisioned throughput does not cause throttling. Sort key is optional in queries filtering by partition key.