AWS Queue-based load levelling architecture

What it runs through

In the order a request or a record moves through it. Each one links to the page that explains it.

Signals in a brief that point here

Phrases that, when you meet them in a scenario, mean this shape rather than another. They are the tell, and they are what an exam question is built around.

• “traffic spike”
• “must never reject an order”
• “downstream can only handle so much”
• “must survive an outage”
• “processed in order”

Learn the services in it

• AWS API GatewayA managed front door for APIs: REST, HTTP and WebSocket APIs that take each call, check who is calling, throttle it, and pass it to Lambda, another service or any HTTP backend.
• AWS LambdaRuns your code in response to events, with no servers to manage. You pay per request and by the millisecond, and a run lasts 15 minutes at most.
• AWS SQSA queue between the part of your system that creates work and the part that does it, so a slow or broken consumer never costs you the work.
• AWS RDS and AuroraManaged relational databases: six engines run for you, plus Aurora, AWS's own MySQL- and PostgreSQL-compatible engine, with standbys, replicas and backups built in.

Shapes it is easily confused with

These run through some of the same services, which is exactly why a brief that fits one can look like it fits another.

• Workflow orchestrationWhen many steps, some slow, and failures must unwind.
• Fan-out to independent consumersWhen one event, several teams, none of them blocking each other.
• Serverless APIWhen ordinary requests, and nothing to patch.