Communication for Software Architects
Communication covers how services talk and coordinate. Synchronous calls are simple but couple availability, asynchronous messaging decouples timing, API gateways and service meshes handle routing, security and observability, and sagas manage business transactions that span services without a single distributed transaction.
Stage: Distributed Systems · Part of the Software Architect roadmap · Last reviewed 19 November 2026
What to learn in Communication
- Choose between synchronous request and asynchronous messaging for a flow
- Describe what an API gateway does and its risks as a bottleneck
- Explain a service mesh sidecar and the features it adds
- Design a saga with compensating actions for an order flow
- Compare orchestration and choreography in sagas
- Set timeouts, retries and backoff to avoid retry storms
Communication interview questions
What is a saga and why is it used?
A saga is a sequence of local transactions across services, where each step has a compensating action to undo it if a later step fails. It provides eventual consistency for business processes when two-phase commit across services is impractical.
When is asynchronous communication better than synchronous?
When the caller does not need an immediate answer, when you want to absorb traffic spikes, or when you want services to keep working if a dependency is down. The cost is more complexity, eventual consistency and harder debugging.
Where Communication fits in the Software Architect roadmap
Communication belongs to the Distributed Systems stage (advanced level), which takes about 7 hours. Reason about failure.
Other units in this stage
Back to the full Software Architect roadmap