Further Reading: Pub/Sub

Back to Pub/Sub: Delivery Guarantees

Pub/Sub Documentation

Official Documentation: Google Cloud Pub/Sub Documentation

Why it matters: Comprehensive official documentation on Pub/Sub architecture, features, and best practices.

Key Concepts

Pub/Sub Architecture: - Topics and subscriptions - Message delivery guarantees - Ordering guarantees

Reliability: - At-least-once delivery - Dead letter queues - Retry policies

Relevance: Provides the authoritative reference for Pub/Sub implementation details.

Google Cloud Architecture Center

Resource: Google Cloud Architecture Center

Why it matters: Reference architectures and best practices for Pub/Sub deployments.

Key Resources

Messaging Patterns: - Event-driven architectures - Microservices communication - Data pipeline patterns

Reliability Patterns: - Idempotency patterns - Retry strategies - Dead letter queue handling

Relevance: Provides real-world architecture examples and best practices.

Additional Resources

Books

"Designing Data-Intensive Applications" by Martin Kleppmann - Chapter on messaging systems - Delivery guarantees and ordering

"Google Cloud Platform in Action" by JJ Geewax - Chapter on Pub/Sub - Pub/Sub examples and best practices

"Site Reliability Engineering" (Google SRE Book) - Chapter on messaging systems - Real-world messaging challenges

Online Resources

Google Cloud Blog: Pub/Sub Articles - Latest Pub/Sub features - Best practices and case studies

GCP Well-Architected Framework: Messaging - Messaging best practices - Design principles

Key Takeaways

  1. At-least-once delivery: Messages may be duplicated, handle idempotently
  2. Ordering is per-key: Ordering only within same ordering key
  3. Dead letter queues: Handle poison messages and failures
  4. Monitor backlog: Track message backlog and processing rate
  5. Design for scale: Pub/Sub scales to millions of messages per second