Unprocessed items handling in DynamoDB - Step-by-Step Execution

Concept Flow - Unprocessed items handling

Send Batch Write Request

↓

Receive Response

↓

Check for Unprocessed Items?

No→Done

Yes↓

Retry Unprocessed Items

↓

Send Batch Write Request

This flow shows how DynamoDB batch write requests may return unprocessed items, which need to be retried until all are processed.

Execution Sample

DynamoDB

batchWrite({
  RequestItems: {
    'Table': [
      { PutRequest: { Item: { /* item attributes */ } } },
      { DeleteRequest: { Key: { /* key attributes */ } } }
    ]
  }
})

Send a batch write request to DynamoDB and handle any unprocessed items by retrying.

Execution Table

Step	Action	Request Items Sent	Response Unprocessed Items	Next Action
1	Send batch write request	5 items	2 items unprocessed	Retry unprocessed items
2	Retry batch write request	2 items	1 item unprocessed	Retry unprocessed items
3	Retry batch write request	1 item	0 items unprocessed	Done - all processed

💡 All items processed when unprocessed items count is zero

Variable Tracker

Variable	Start	After 1	After 2	Final
UnprocessedItemsCount	5	2	1	0

Key Moments - 2 Insights

Why do some items remain unprocessed after the first batch write?

Because DynamoDB may throttle requests or have temporary capacity limits, causing some items to be returned as unprocessed (see execution_table step 1).

Do we stop retrying after the first retry?

No, we keep retrying unprocessed items until none remain, as shown in execution_table steps 2 and 3.

Visual Quiz - 3 Questions

Look at the execution table, how many items were unprocessed after the first batch write?

A 1 item

B 5 items

C 2 items

D 0 items

💡 Hint

Check the 'Response Unprocessed Items' column in step 1 of the execution_table.

At which step does the unprocessed items count become zero?

A Step 3

B Step 2

C Step 1

D Never

💡 Hint

Look at the 'UnprocessedItemsCount' variable in variable_tracker after each step.

If the first batch write had no unprocessed items, what would happen next?

A Retry unprocessed items

B Finish processing

C Send another batch write with same items

D Throw an error

💡 Hint

Refer to the concept_flow where 'No' from 'Check for Unprocessed Items?' leads to 'Done'.

Concept Snapshot

DynamoDB batch writes may return unprocessed items due to throttling.
Always check response for unprocessed items.
Retry only those unprocessed items until all succeed.
This ensures data consistency and reliability.

Full Transcript

When you send a batch write request to DynamoDB, it may not process all items immediately. Some items can be returned as unprocessed due to temporary limits. You must check the response for these unprocessed items. If there are any, retry sending only those items. Repeat this process until no unprocessed items remain. This way, you ensure all your data is written successfully.