Protect SLA and Choose Storage
Company: DoorDash
Role: Software Engineer
Category: System Design
Difficulty: medium
Interview Round: Technical Screen
You own a service that calculates delivery-order payments from order lifecycle events.
Discuss two design questions:
1. If an upstream dependency suddenly has very high latency, how would you keep your own service within its SLA?
2. For storing order state and payment data, how would you choose between Redis, a relational database, or another storage system? Explain the trade-offs and your recommended architecture.
Overview: This question evaluates system design competencies in service reliability and storage architecture, focusing on SLA protection during upstream latency spikes and trade-offs in choosing in-memory caches (e.g., Redis), relational databases, or alternative storage for order and payment state.