Parece simples até o entrevistador perguntar "why?". Abaixo, uma estrutura de resposta que dá conta do follow-up.
A managed function runtime that runs code in response to invocations and events. Billing and capacity behavior depend on the execution mode and configuration, rather than only counting executions.
A cold start adds initialization work when a new execution environment is needed. Package size, initialization, runtime, and dependencies influence it. Provisioned concurrency can reduce cold-start exposure for configured capacity, with an additional cost.
The default maximum invocation timeout is 15 minutes. Lambda Managed Instances allow up to 90 minutes for asynchronous invocations and supported event source mappings, excluding Amazon MQ and Amazon DocumentDB. Synchronous invocations remain limited to 15 minutes. Check the execution mode and invocation type before designing around a limit.
Use functions for work that fits their duration, resource, and concurrency constraints. Persistent connection servers or longer processing may need containers or an orchestrated multi-step design. A timeout is not proof that no external effect occurred.
Sources: AWS timeout limits and Managed Instances execution behavior.
Ler é o primeiro passo. No app, essa pergunta vira flashcard com repetição espaçada, e o módulo tem quiz e simulado para testar o que ficou.
Treinar essa pergunta no app