Option B is preferred over A because Amazon Kinesis Data Streams inherently maintain the order of records within a shard, which is crucial for the given requirement of preserving the order of messages for a particular payment ID. When you use the payment ID as the partition key, all messages for that payment ID will be sent to the same shard, ensuring that the order of messages is maintained. On the other hand, Amazon DynamoDB is a NoSQL database service that provides fast and predictable performance with seamless scalability. While it can store data with partition keys, it does not guarantee the order of records within a partition, which is essential for the given use case. Hence, using Kinesis Data Streams is more suitable for this requirement. As DynamoDB does not keep the order, I think BE is the correct answer here.

I don't understand the question. The only requirement is: " system that requires messages for a particular payment ID to be received in the same order that they were sent" SQS FIFO (E) meets this requirement. Why would you "write the message" to Kinesis or DynamoDB anymore. There is no streaming or DB storage requirement in the question. Between A/B, B is better logically but it doesn't meet any stated requirement. Happy to understand what I'm missing

Both Kinesis and SQS FIFO queue guarantee the order, other answers don't.

Option B (Write the messages to an Amazon Kinesis data stream with the payment ID as the partition key): Kinesis can provide ordered processing within a shard Write the messages to an Amazon Simple Queue Service (Amazon SQS) FIFO queue. Set the message group to use the payment ID. SQS FIFO (First-In-First-Out) queues preserve the order of messages within a message group.

Technically both B and E will ensure processing order, but SQS FIFO was specifically built to handle this requirement. There is no ask on how to store the data so A and C are out.

https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.Partitions.html

options D and E are better because they mimic a real-world queue system and ensure that payments are processed in the correct order, just like customers in a store would be served in the order they arrived. This is crucial for a payment processing system where order matters to avoid mistakes in payment processing.

AAAAAAAAA EEEEEEEEEEEEEE

IF the question would be "Choose all the solutions that fulfill these requirements" I would chosen BE. But it is: "Which actions should a solutions architect take to meet this requirement? " For this reason I chose AE, because we don't need both Kinesis AND SQS for this solution. Both choices complement to order processing: order stored in DB, work item goes to the queue.

E --> no doubt B --> see https://docs.aws.amazon.com/streams/latest/dev/key-concepts.html

1) SQS FIFO queues guarantee that messages are received in the exact order they are sent. Using the payment ID as the message group ensures all messages for a payment ID are received sequentially. 2) Kinesis data streams can also enforce ordering on a per partition key basis. Using the payment ID as the partition key will ensure strict ordering of messages for each payment ID.

BE no doubt.

Option A, writing the messages to an Amazon DynamoDB table, would not necessarily preserve the order of messages for a particular payment ID

I don´t unsderstand A, How you can guaratee the order with DynamoDB?? The order is guarantee with SQS FIFO and Kinesis Data Stream in 1 shard...

AE is the answer

dynamodb or kinesis data stream which one in order?

No doubt )