Skip to main contentIBM Automation Event-Driven Reference Architecture

Generic Business Use Cases

The adoption of event-driven architecture is based four main technical drivers:

4 usecases
  • Communication layer:

    • Adopt messaging and asynchronous communication between applications and microservices for loosely coupled services (See this note where we present a way to support a service mesh solution using asynchronous event). The data are pushed as immutables record, or commands are sent with exactly once delivery. Reducing pull data approach.
    • Pub/sub messaging for cloud native applications to improve communication inter microservices.
    • Support Always-on services with asynchronous data replication.
    • Expose data to any application to consume.
  • Reactive systems: to support reactive, responsive applications for addressing resiliency, scaling so adopting message buses.

  • Data Agility:

    • Centralize online data pipeline to decouple applications and microservices.
    • Monitor distributed applications to produce centralized feed of operational data.
    • Logs collector from multiple services.
    • Implement event sourcing pattern out of the box, using configuration to keep message for a long time period. Data are replicated between brokers within the cluster and cross availability zones if needed.
  • Near real-time analytics insights:

    • Aggregation of event coming from multiple producers.
    • Look at event sequencing patterns
    • Compute next best action from event streams