Overview

  • Global message bus—consistent latency globally
  • Scalable, durable
  • Topic-based
  • At-least-once model
  • Push and pull modes
  • Exactly once processing
  • Messages can be ordered and filtered