BatchWriteItem in DynamoDB - Time & Space Complexity

Time Complexity: BatchWriteItem

O(n)

Understanding Time Complexity

When using BatchWriteItem in DynamoDB, it's important to understand how the time it takes grows as you add more items to write.

We want to know: how does the execution time change when we increase the number of items in the batch?

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

const params = {
  RequestItems: {
    'MyTable': [
      { PutRequest: { Item: { id: '1', data: 'A' } } },
      { PutRequest: { Item: { id: '2', data: 'B' } } },
      // ... up to 25 items
    ]
  }
};

const result = await dynamodb.batchWrite(params).promise();
    

This code sends a batch write request to DynamoDB to insert up to 25 items in one call.

Identify Repeating Operations

Identify the loops, recursion, array traversals that repeat.

• Primary operation: Writing each item in the batch to the database.
• How many times: Once per item in the batch, up to 25 items per request.

How Execution Grows With Input

As you add more items to the batch, the time to process grows roughly in direct proportion to the number of items.

Input Size (n)	Approx. Operations
10	About 10 write operations
25	About 25 write operations (maximum per batch)
50	Two batches of 25 writes each, so about 50 write operations

Pattern observation: The time grows linearly with the number of items, but each batch can only hold up to 25 items.

Final Time Complexity

Time Complexity: O(n)

This means the time to complete the batch write grows directly with the number of items you want to write.

Common Mistake

[X] Wrong: "BatchWriteItem writes all items instantly regardless of batch size."

[OK] Correct: Each item still requires a write operation, so more items mean more work and more time.

Interview Connect

Understanding how batch operations scale helps you design efficient database interactions and shows you can think about performance in real applications.

Self-Check

"What if we split a batch of 50 items into two parallel BatchWriteItem calls? How would the time complexity change?"

Practice

1.

What is the main purpose of the BatchWriteItem operation in DynamoDB?

easy

A. To create a new table in DynamoDB

B. To read multiple items from different tables

C. To add or delete multiple items in a single request

D. To update a single item with new attributes

Solution

1. Step 1: Understand BatchWriteItem functionality

   The BatchWriteItem operation allows you to write (add or delete) multiple items in one request to save time and reduce network calls.

2. Step 2: Compare options with BatchWriteItem purpose

   Only To add or delete multiple items in a single request correctly describes adding or deleting multiple items in one request, which matches BatchWriteItem's purpose.

3. Final Answer:

   To add or delete multiple items in a single request -> Option C

4. Quick Check:

   BatchWriteItem = add/delete many items [OK]

Hint: BatchWriteItem = batch add or delete items [OK]

Common Mistakes:

• Confusing BatchWriteItem with BatchGetItem
• Thinking it updates items instead of add/delete
• Assuming it creates tables

2.

Which of the following is the correct structure for a BatchWriteItem request in DynamoDB?

{
  "RequestItems": {
    "TableName": [
      { "PutRequest": { "Item": { ... } } },
      { "DeleteRequest": { "Key": { ... } } }
    ]
  }
}

easy

A. RequestItems is an array of table names with PutRequest and DeleteRequest inside

B. RequestItems contains table names as keys and arrays of PutRequest or DeleteRequest objects as values

C. RequestItems is a list of items without specifying table names

D. RequestItems only accepts PutRequest objects, not DeleteRequest

Solution

1. Step 1: Review BatchWriteItem request format

   The RequestItems key maps table names to arrays of write requests, each being either PutRequest or DeleteRequest.

2. Step 2: Match options to correct format

   RequestItems contains table names as keys and arrays of PutRequest or DeleteRequest objects as values correctly describes the structure: table names as keys, arrays of PutRequest/DeleteRequest as values.

3. Final Answer:

   RequestItems contains table names as keys and arrays of PutRequest or DeleteRequest objects as values -> Option B

4. Quick Check:

   RequestItems = table name keys + Put/Delete arrays [OK]

Hint: RequestItems keys are table names with Put/Delete arrays [OK]

Common Mistakes:

• Using array of table names instead of object with table keys
• Omitting DeleteRequest option
• Not nesting PutRequest/DeleteRequest inside arrays

3.

Given the following BatchWriteItem request, what will be the result if one item fails due to exceeding provisioned throughput?

{
  "RequestItems": {
    "Products": [
      { "PutRequest": { "Item": { "ID": { "S": "101" }, "Name": { "S": "Pen" } } } },
      { "PutRequest": { "Item": { "ID": { "S": "102" }, "Name": { "S": "Notebook" } } } }
    ]
  }
}

medium

A. The entire batch fails and no items are written

B. The request throws a syntax error

C. Both items are written regardless of failure

D. Only the failed item is returned in UnprocessedItems for retry

Solution

1. Step 1: Understand BatchWriteItem error handling

   If some items exceed provisioned throughput, DynamoDB returns those items in UnprocessedItems for retry, not failing the entire batch.

2. Step 2: Analyze the given scenario

   One item fails due to throughput, so only that item is unprocessed and returned for retry, while others succeed.

3. Final Answer:

   Only the failed item is returned in UnprocessedItems for retry -> Option D

4. Quick Check:

   UnprocessedItems = failed items for retry [OK]

Hint: Failed items appear in UnprocessedItems for retry [OK]

Common Mistakes:

• Assuming all items fail if one fails
• Thinking BatchWriteItem throws syntax error on throughput issues
• Believing all items succeed regardless of errors

4.

Identify the error in this BatchWriteItem request snippet:

{
  "RequestItems": {
    "Orders": [
      { "PutRequest": { "Item": { "OrderId": "123", "Amount": 50 } } }
    ]
  }
}

medium

A. The attribute values are missing DynamoDB data types

B. The table name should be an array, not an object

C. PutRequest should be outside the array

D. BatchWriteItem cannot write to Orders table

Solution

1. Step 1: Check attribute value format in BatchWriteItem

   DynamoDB requires attribute values to specify data types like {"S": "string"} or {"N": "number"} in BatchWriteItem requests.

2. Step 2: Identify the error in the snippet

   Attributes "OrderId" and "Amount" lack data type wrappers, so the request is invalid.

3. Final Answer:

   The attribute values are missing DynamoDB data types -> Option A

4. Quick Check:

   Attribute values need data types like {"S": "text"} [OK]

Hint: Attribute values must include DynamoDB data types [OK]

Common Mistakes:

• Omitting data types for attribute values
• Misunderstanding table name structure
• Placing PutRequest incorrectly

5.

You want to write 50 items to a DynamoDB table using BatchWriteItem. Since the API limits 25 items per batch, how should you handle this in your application?

hard

A. Split the 50 items into two batches of 25 and retry unprocessed items after each batch

B. Send all 50 items in one BatchWriteItem request and expect it to succeed

C. Write items one by one using PutItem to avoid batch limits

D. Use BatchGetItem instead to write multiple items

Solution

1. Step 1: Recognize BatchWriteItem limits

   BatchWriteItem supports a maximum of 25 items per request, so 50 items must be split into multiple batches.

2. Step 2: Implement batching and retry logic

   Split 50 items into two batches of 25 each. After each batch, check UnprocessedItems and retry those to ensure all items are written.

3. Final Answer:

   Split the 50 items into two batches of 25 and retry unprocessed items after each batch -> Option A

4. Quick Check:

   BatchWriteItem max 25 items; split and retry [OK]

Hint: Split items into max 25 per batch and retry unprocessed [OK]

Common Mistakes:

• Sending more than 25 items in one batch
• Ignoring unprocessed items and not retrying
• Using BatchGetItem to write items