Choosing the Right Messaging Solution for Your Architecture
summary
Compares AWS SQS, Azure Service Bus and RabbitMQ with concrete tradeoffs: SQS gives 256 KB limit (workaround via S3), FIFO queues for exactly-once + ordering, 10-message batching, long polling up to 20s and visibility timeout 30s–12h; Azure Service Bus offers peek-lock (non-destructive) mode, premium tier up to 100 MB messages, automatic DLQs, sessions for ordered processing and 2,000 subscriptions, plus transactional send/complete semantics; RabbitMQ exposes exchanges/bindings, supports AMQP, prefetch=1 fair dispatch, priority and quorum queues but requires ops for clustering and durability.
AI-written summary · about
chapters (18)
- 0:00Messaging architecture choices
- 1:51Messaging vs synchronous HTTP
- 4:00Popular message brokers
- 10:27SQS send/receive patterns
- 15:47SQS polling and visibility timeout
- 21:30SQS FIFO queues and ordering
- 22:12SQS publish/subscribe with SNS
- 26:52Azure Service Bus enterprise broker
- 30:02Azure Service Bus send/receive modes
- 33:31Azure Service Bus publish/subscribe
- 38:46Azure Service Bus dead letter queues
- 40:12Azure Service Bus message sessions
- 43:43Azure Service Bus transactionality
- 47:18RabbitMQ overview and setup
- 50:25RabbitMQ competing consumers
- 52:59RabbitMQ exchanges and bindings
- 57:16RabbitMQ advanced exchange types
- 1:00:47Choosing the right solution
related talks
01h 16mChoosing the Right Messaging Solution for Your Architecture
50mMiddleware Showdown: Exploring Diverse Messaging Solutions
53mAzure Messaging, Queueing, and Events - Which service is right for your solution?
55mImplementing scalable messaging patterns in Azure
30mApplication integration patterns for microservices
46mReal-world Examples of Message Queue Usage: Insights from RabbitMQ
42mIt's like 10,000 streams when what you need is a queue
01h 01mBeyond Pub/Sub - Advanced Messaging Patterns