Basic scan operation in DynamoDB - Step-by-Step Execution

Concept Flow - Basic scan operation

Start Scan Request

↓

Read First Page of Items

↓

Return Items to User

↓

Check for More Pages?

No→End Scan

Yes↓

Read Next Page of Items

↩Back to Return Items

The scan operation reads all items in a table page by page, returning them until no more pages remain.

Execution Sample

DynamoDB

ScanRequest = {'TableName': 'Books'}
response = dynamodb.scan(**ScanRequest)
items = response['Items']
while 'LastEvaluatedKey' in response:
  response = dynamodb.scan(TableName='Books', ExclusiveStartKey=response['LastEvaluatedKey'])
  items.extend(response['Items'])

This code scans the entire 'Books' table, fetching all items page by page.

Execution Table

Step	Action	Page Number	Items Returned	LastEvaluatedKey Present	Next Step
1	Send scan request to 'Books' table	1	5 items	Yes	Fetch next page
2	Send scan request with ExclusiveStartKey	2	5 items	Yes	Fetch next page
3	Send scan request with ExclusiveStartKey	3	3 items	No	End scan
4	No more pages	-	0 items	No	Stop

💡 No LastEvaluatedKey after page 3 means all items have been scanned.

Variable Tracker

Variable	Start	After Step 1	After Step 2	After Step 3	Final
response.Items	empty	5 items	5 items	3 items	13 items total
response.LastEvaluatedKey	none	present	present	none	none
items	empty list	5 items	10 items	13 items	13 items

Key Moments - 2 Insights

Why do we need to check for LastEvaluatedKey after each scan?

Because DynamoDB returns data in pages, LastEvaluatedKey tells us if more pages exist. Without checking it (see execution_table rows 1-3), we might miss items.

What happens if we ignore LastEvaluatedKey and only scan once?

We only get the first page of items (row 1), missing the rest of the table data.

Visual Quiz - 3 Questions

Look at the execution table, how many items are returned after step 2?

A 10 items

B 3 items

C 5 items

D 13 items

💡 Hint

Check the 'Items Returned' column for step 2 and the 'items' variable in variable_tracker after step 2.

At which step does the scan operation know it has reached the last page?

A Step 1

B Step 2

C Step 3

D Step 4

💡 Hint

Look at the 'LastEvaluatedKey Present' column in execution_table; it is 'No' at step 3.

If the table had 20 items and each page returned 5 items, how many steps would the scan take?

A 3 steps

B 4 steps

C 5 steps

D 6 steps

💡 Hint

Each step returns 5 items; 20 items total means 4 full pages plus no more pages after the 4th.

Concept Snapshot

Basic Scan Operation in DynamoDB:
- Reads all items in a table, page by page.
- Each scan returns a page of items and a LastEvaluatedKey if more pages exist.
- Use LastEvaluatedKey to fetch next page.
- Repeat until no LastEvaluatedKey.
- Returns all items but can be costly for large tables.

Full Transcript

The Basic Scan Operation in DynamoDB reads all items from a table by fetching pages of data. Each scan request returns some items and a LastEvaluatedKey if there are more pages. The client checks this key to continue scanning until all items are retrieved. This process is shown step-by-step in the execution table, where each step fetches a page of items. Variables like response.Items and response.LastEvaluatedKey change after each scan. Beginners often wonder why checking LastEvaluatedKey is necessary; it ensures no items are missed. The visual quiz tests understanding of how many items are returned at each step and when the scan ends.

Practice

1. What does the scan operation do in DynamoDB?

easy

A. Reads all items in a table

B. Reads only one item by key

C. Deletes items from a table

D. Updates items in a table

Solution

1. Step 1: Understand the scan operation

   The scan operation reads every item in the DynamoDB table without filtering by key.

2. Step 2: Compare with other operations

   Unlike get or query, scan reads all items, not just specific keys.

3. Final Answer:

   Reads all items in a table -> Option A

4. Quick Check:

   Scan = Reads all items [OK]

Hint: Scan reads entire table, not just keys [OK]

Common Mistakes:

• Confusing scan with get or query
• Thinking scan deletes or updates data
• Assuming scan reads only filtered items

2. Which of the following is the correct syntax to perform a scan operation using AWS SDK for JavaScript v3?

easy

A. const data = await client.get({ TableName: 'MyTable' });

B. const data = await client.query({ TableName: 'MyTable' });

C. const data = await client.scan({ TableName: 'MyTable' });

D. const data = await client.delete({ TableName: 'MyTable' });

Solution

1. Step 1: Identify scan method usage

   The scan method is called on the DynamoDB client with parameters including TableName.

2. Step 2: Check other methods

   Get, query, and delete are different operations and do not perform scan.

3. Final Answer:

   const data = await client.scan({ TableName: 'MyTable' }); -> Option C

4. Quick Check:

   Scan syntax uses client.scan() [OK]

Hint: Scan uses client.scan() with TableName [OK]

Common Mistakes:

• Using get or query instead of scan
• Missing await keyword
• Wrong method names like delete

3. Given a DynamoDB table with 3 items: {id:1, name:'A'}, {id:2, name:'B'}, {id:3, name:'C'}, what will the scan operation return?

medium

A. [{id:1, name:'A'}, {id:2, name:'B'}, {id:3, name:'C'}]

B. Error: No items found

C. []

D. [{id:1, name:'A'}]

Solution

1. Step 1: Understand scan returns all items

   Scan reads every item in the table, so all 3 items will be returned.

2. Step 2: Check options for completeness

   Only [{id:1, name:'A'}, {id:2, name:'B'}, {id:3, name:'C'}] lists all 3 items; others are incomplete or errors.

3. Final Answer:

   [{id:1, name:'A'}, {id:2, name:'B'}, {id:3, name:'C'}] -> Option A

4. Quick Check:

   Scan returns all items [OK]

Hint: Scan returns full table items list [OK]

Common Mistakes:

• Expecting scan to return only one item
• Thinking scan returns empty if no filter
• Confusing scan with query results

4. You wrote this code to scan a DynamoDB table but get no results:

const params = { TableName: 'MyTable', FilterExpression: 'age > :val', ExpressionAttributeValues: { ':val': 30 } };
const data = await client.scan(params);

What is the likely problem?

medium

A. TableName is missing in params

B. FilterExpression syntax is incorrect, causing scan to fail

C. Scan does not support FilterExpression

D. FilterExpression is applied after scan reads all items, so no items match age > 30

Solution

1. Step 1: Understand FilterExpression in scan

   FilterExpression filters results after scanning all items; if no items match, result is empty.

2. Step 2: Check syntax and params

   Syntax is correct, TableName is present, and scan supports FilterExpression.

3. Final Answer:

   FilterExpression is applied after scan reads all items, so no items match age > 30 -> Option D

4. Quick Check:

   FilterExpression filters after scan [OK]

Hint: FilterExpression filters after scan reads all items [OK]

Common Mistakes:

• Thinking FilterExpression prevents scanning items
• Assuming scan fails with FilterExpression
• Missing TableName parameter

5. You want to scan a large DynamoDB table but only retrieve items where status is 'active'. Which approach is best to reduce data returned and improve performance?

hard

A. Use scan with FilterExpression 'status = :s' and ExpressionAttributeValues { ':s': 'active' }

B. Use query operation with status as partition key

C. Use scan without filters and filter results in application code

D. Use scan with ProjectionExpression to get only 'status' attribute

Solution

1. Step 1: Understand scan vs query

   Scan reads entire table; query reads items by key, more efficient for filtering.

2. Step 2: Check if status can be partition key

   If status is partition key, query can efficiently get only 'active' items without scanning all.

3. Step 3: Evaluate other options

   FilterExpression filters after scan, so less efficient; filtering in app wastes bandwidth; ProjectionExpression only limits attributes, not items.

4. Final Answer:

   Use query operation with status as partition key -> Option B

5. Quick Check:

   Query with key filters efficiently [OK]

Hint: Query by key is faster than scan with filters [OK]

Common Mistakes:

• Relying on scan with filters for large tables
• Filtering data in application instead of query
• Confusing ProjectionExpression with filtering items