Examkin

Question

Difficulty: Very hardDecoupling Architectures and Event-Driven Messaging

An enterprise retail organization is migrating its inventory management pipeline to AWS. The pipeline must process inventory state-change events generated by retail stores. The requirements are:

1. Event processing must be strictly ordered on a per-store basis to prevent race conditions in warehouse stock levels.
2. The events must be fanned out to two separate downstream microservices: a real-time inventory reconciliation service and a near-real-time business intelligence dashboard.
3. The total ingestion throughput across all stores is expected to exceed 15,000 messages per second, while individual store volume varies dynamically.

Which TWO configurations should a solutions architect combine to meet these requirements with the lowest operational overhead? (Select TWO.)

  1. Create an Amazon SNS FIFO topic and subscribe two Amazon SQS FIFO queues to the topic, setting the retail store ID as the Message Group ID for all published events.Answer
  2. Enable high-throughput FIFO for both the Amazon SNS FIFO topic and the subscribed Amazon SQS FIFO queues to support the required message throughput.Answer
  3. C
    Configure an Amazon Kinesis Data Stream with the retail store ID as the partition key, and configure two independent consumers to process the stream.
  4. D
    Create an Amazon SNS Standard topic with two Amazon SQS Standard queues as subscribers, and rely on downstream services to re-sequence events using a custom timestamp header.
  5. E
    Ingest all events into a single Amazon SQS Standard queue first to handle high volume, and use an AWS Lambda function to route messages to separate SQS FIFO queues based on the store ID.

Answer

The correct solution involves creating an Amazon SNS FIFO topic that fans out to two Amazon SQS FIFO queues, using the store ID as the Message Group ID, and enabling high-throughput FIFO mode on both the SNS topic and the SQS queues.
The correct solution uses an Amazon SNS FIFO topic to fan out inventory events to two Amazon SQS FIFO queues (one for each downstream service). By setting the store ID as the Message Group ID, messages are processed in order for each individual store while allowing parallel processing across different stores. To handle the 15,000 messages per second throughput, high-throughput FIFO mode must be enabled on both the SNS FIFO topic and the SQS FIFO queues, which scales the throughput limits.

Step-by-Step Solution

1
Analyze the ordering and fan-out requirements.
Identify that the system requires message ordering grouped by store ID and delivery to two independent downstream services.
Ordering requires FIFO capability (using a Message Group ID or partition key), and delivery to two independent services requires a pub/sub fan-out pattern.
2
Evaluate the throughput requirements against standard AWS service limits.
Identify that the total throughput of 15,000 messages/sec exceeds the default limit of SQS/SNS FIFO queues (which is 300 messages/sec without high-throughput mode).
High-throughput FIFO mode must be enabled to scale SQS and SNS FIFO to support 15,000 messages/sec.
3
Determine the configuration with the lowest operational overhead.
Select SNS FIFO subscribed to SQS FIFO with high-throughput mode enabled, rather than provisioning and managing Kinesis shards.
SNS/SQS is fully serverless and handles scaling automatically without manual shard management, minimizing operational overhead.

Key Concept

Decoupling event-driven architectures with high-throughput SNS FIFO and SQS FIFO to maintain ordering within specific message groups.
Rate this question

Topics

Decoupling Architectures and Event-Driven Messaging

More questions from this topic