One-to-many relationship patterns in DynamoDB

Introduction

One-to-many relationships help you connect one item to many related items in your database. This lets you organize data like a list of orders for one customer.

You want to store all orders made by a single customer.

You need to keep track of all comments on a blog post.

You want to list all products in a category.

You want to save all messages in a chat room.

You want to group all tasks under one project.

Syntax

DynamoDB

PartitionKey: unique ID for the main item (e.g., CustomerID)
SortKey: unique ID for each related item (e.g., OrderID)
Attributes: other details about each item

Use the PartitionKey to group related items together.

Use the SortKey to order or identify each related item uniquely.

Examples

This stores an order for customer 123 with order ID 001.

DynamoDB

PartitionKey: CUSTOMER#123
SortKey: ORDER#001
Attributes: {OrderDate: '2024-06-01', Total: 50}

This stores another order for the same customer with a different order ID.

DynamoDB

PartitionKey: CUSTOMER#123
SortKey: ORDER#002
Attributes: {OrderDate: '2024-06-02', Total: 75}

This stores a comment on a blog post with ID 456.

DynamoDB

PartitionKey: POST#456
SortKey: COMMENT#001
Attributes: {User: 'Alice', Text: 'Nice post!'}

Sample Program

This example creates a table to store orders. It inserts two orders for one customer and then retrieves all orders for that customer.

DynamoDB

CREATE TABLE Orders (
  PartitionKey STRING,
  SortKey STRING,
  OrderDate STRING,
  Total NUMBER,
  PRIMARY KEY (PartitionKey, SortKey)
);

-- Insert two orders for customer 123
INSERT INTO Orders (PartitionKey, SortKey, OrderDate, Total) VALUES ('CUSTOMER#123', 'ORDER#001', '2024-06-01', 50);
INSERT INTO Orders (PartitionKey, SortKey, OrderDate, Total) VALUES ('CUSTOMER#123', 'ORDER#002', '2024-06-02', 75);

-- Query all orders for customer 123
SELECT SortKey, OrderDate, Total FROM Orders WHERE PartitionKey = 'CUSTOMER#123' ORDER BY SortKey;

Important Notes

Use clear prefixes like CUSTOMER# or ORDER# to keep keys readable and organized.

Sorting by SortKey helps you get related items in order, like by date or ID.

Remember DynamoDB tables need a PartitionKey and SortKey for one-to-many patterns.

Summary

One-to-many relationships group many related items under one main item using PartitionKey and SortKey.

Use PartitionKey to identify the main item and SortKey to identify each related item.

This pattern helps organize data like orders per customer or comments per post.

Practice

1. What is the main purpose of using a PartitionKey in a one-to-many relationship pattern in DynamoDB?

easy

A. To group all related items under one main item

B. To uniquely identify each related item

C. To store the data type of the item

D. To define the size of the table

Solution

1. Step 1: Understand PartitionKey role

   The PartitionKey groups related items together in DynamoDB, acting like a folder for related data.

2. Step 2: Differentiate PartitionKey from SortKey

   The SortKey uniquely identifies each item within the group, but PartitionKey groups them all.

3. Final Answer:

   To group all related items under one main item -> Option A

4. Quick Check:

   PartitionKey groups related items = D [OK]

Hint: PartitionKey groups items; SortKey identifies each item [OK]

Common Mistakes:

• Confusing PartitionKey with SortKey
• Thinking PartitionKey uniquely identifies each item
• Assuming PartitionKey stores data type

2. Which of the following is the correct way to define a SortKey in a DynamoDB table for a one-to-many relationship?

easy

A. SortKey as the PartitionKey

B. SortKey as the same value for all items

C. SortKey as a unique identifier for each related item

D. SortKey as the table name

Solution

1. Step 1: Understand SortKey purpose

   The SortKey uniquely identifies each related item within the same PartitionKey group.

2. Step 2: Eliminate incorrect options

   SortKey cannot be the same for all items or equal to PartitionKey or table name.

3. Final Answer:

   SortKey as a unique identifier for each related item -> Option C

4. Quick Check:

   SortKey uniquely identifies items = C [OK]

Hint: SortKey uniquely identifies related items within a PartitionKey [OK]

Common Mistakes:

• Using same SortKey for all items
• Confusing SortKey with PartitionKey
• Assigning table name as SortKey

3. Given a DynamoDB table with PartitionKey = 'CustomerID' and SortKey = 'OrderID', what will the query Query(CustomerID='123') return?

medium

A. Only the order with OrderID 123

B. All orders for customer with ID 123

C. All customers with OrderID 123

D. An error because SortKey is missing

Solution

1. Step 1: Understand Query with PartitionKey only

   Querying with PartitionKey='123' returns all items sharing that PartitionKey, i.e., all orders for customer 123.

2. Step 2: Clarify SortKey role in query

   SortKey is not specified, so all SortKey values under PartitionKey='123' are returned.

3. Final Answer:

   All orders for customer with ID 123 -> Option B

4. Quick Check:

   Query by PartitionKey returns all related items = A [OK]

Hint: Query by PartitionKey returns all related items [OK]

Common Mistakes:

• Thinking query returns single item without SortKey
• Confusing PartitionKey and SortKey in query
• Expecting error when SortKey is not specified

4. You have a DynamoDB table with PartitionKey = 'PostID' and SortKey = 'CommentID'. You try to insert two comments with the same PostID and CommentID. What is the issue?

medium

A. The second insert overwrites the first comment

B. Both comments are stored separately

C. DynamoDB throws a syntax error

D. PartitionKey must be unique, so insert fails

Solution

1. Step 1: Understand uniqueness of PartitionKey + SortKey

   In DynamoDB, the combination of PartitionKey and SortKey must be unique for each item.

2. Step 2: Analyze inserting duplicate keys

   Inserting an item with the same PartitionKey and SortKey overwrites the existing item.

3. Final Answer:

   The second insert overwrites the first comment -> Option A

4. Quick Check:

   Duplicate keys overwrite existing item = A [OK]

Hint: Duplicate PartitionKey+SortKey overwrites existing item [OK]

Common Mistakes:

• Expecting error on duplicate keys
• Thinking both items are stored separately
• Confusing PartitionKey uniqueness with SortKey

5. You want to model a one-to-many relationship where each customer can have multiple orders, and each order can have multiple items. Which DynamoDB design pattern best fits this scenario?

hard

A. Use PartitionKey as CustomerID and SortKey as OrderID#ItemID

B. Use PartitionKey as OrderID and SortKey as CustomerID#ItemID

C. Use PartitionKey as ItemID and SortKey as CustomerID#OrderID

D. Use PartitionKey as CustomerID#OrderID and SortKey as ItemID

Solution

1. Step 1: Understand the hierarchy

   Customer has many orders, and each order has many items, so two levels of one-to-many.

2. Step 2: Design composite keys for nested relationships

   Using PartitionKey as CustomerID#OrderID groups items by order under each customer; SortKey as ItemID identifies each item.

3. Step 3: Eliminate other options

   Other options mix keys incorrectly, losing the hierarchical grouping.

4. Final Answer:

   Use PartitionKey as CustomerID#OrderID and SortKey as ItemID -> Option D

5. Quick Check:

   Composite PartitionKey groups nested one-to-many = B [OK]

Hint: Use composite PartitionKey for nested one-to-many groups [OK]

Common Mistakes:

• Using wrong key order losing hierarchy
• Not combining keys for nested relationships
• Confusing PartitionKey and SortKey roles