The Rise of Reactive Microservices

Are your microservices just a distributed monolith? Learn how to build truly resilient systems by rethinking service boundaries.

#1about 3 minutes

The pitfalls of synchronous microservice communication

Synchronous calls between services create tight coupling and lead to cascading failures when downstream systems experience latency.

#2about 4 minutes

Decoupling services with asynchronous message queues

Using message queues like RabbitMQ isolates high-latency services and improves system resilience by changing from immediate creation to eventual acceptance.

#3about 4 minutes

Avoiding the entity service anti-pattern in architecture

Slicing microservices based on data entities like 'user' or 'account' creates dependencies across teams and slows down feature development.

#4about 7 minutes

Using push-based streams for data synchronization

Reactive systems push data changes through event streams, allowing services to build local read models (projections) and achieve runtime autonomy.

#5about 7 minutes

Navigating the challenges of eventual consistency

Eventual consistency involves three main challenges—divergence, variance, and latency—which can be managed with ordering, deduplication, and optimistic UIs.

#6about 5 minutes

The key benefits of building reactive systems

Reactive architectures improve resilience and enable horizontal scaling by creating autonomous services that can build real-time, push-based user interfaces.

#7about 2 minutes

Decomposing monoliths with change data capture

Change Data Capture (CDC) streams events directly from a legacy database's transaction log, enabling gradual monolith decomposition without modifying the legacy code.

#8about 6 minutes

Achieving massive throughput with sharded architectures

Sharding partitions data and processing across independent streams, enabling systems to scale horizontally and handle nearly unlimited throughput.

#9about 3 minutes

Balancing architectural complexity and business needs

Evolving from a monolith to a sharded, stream-based system adds powerful capabilities but also increases complexity, so always choose the simplest architecture that meets your requirements.

#10about 3 minutes

Answering questions on micro-frontends and data ownership

The speaker discusses how micro-frontends align with team autonomy and clarifies that while data can be copied to projections, updates must go through the owning service.