Design Amazon

Interview-Ready Approach

1) Clarify Scope and SLOs

• •Problem statement: Design Amazon's e-commerce platform - product catalog, cart, checkout, recommendations, and warehouse fulfillment at 300 M+ users.
• •Design for a peak load target around 20,000 RPS (including burst headroom).
• •Active customers: 300,000,000+
• •Products (total): 350,000,000+
• •Peak orders/second: ~10,000
• •Search QPS (peak): ~50,000
• •Search latency (P99): < 300 ms

2) Capacity Planning Method

• •Convert traffic and growth constraints into request rate, storage growth, and concurrency budgets.
• •Keep at least 2-3x safety margin per tier (ingress, compute, storage, async workers).
• •Reserve explicit latency budgets per hop so p95 can be defended in review.

3) Architecture Decisions

• •Databases: Define a clear system-of-record and design read/write paths separately before adding optimizations.
• •Caching: Put cache on hot read paths first and pick cache-aside or write-through explicitly.
• •Microservices: Split services by business boundary, not by technical layer, and enforce ownership per domain.
• •Search: Use primary store for writes and async index updates for search relevance + scale.
• •Message Queues: Move non-blocking and retry-heavy work to async consumers with explicit retry and DLQ policies.
• •Sharding: Choose shard keys around access patterns and growth hotspots, not just data size.

4) Reliability and Failure Strategy

• •Use strong write constraints (transactions or conditional writes) and explicit backup/restore strategy.
• •Bound staleness with TTL + invalidation hooks for critical entities.
• •Add service-level timeout/retry budgets and contract tests.
• •Track indexing lag and support reindex from source of truth.
• •Guarantee idempotent consumers and trace every message with correlation IDs.

5) Validation Plan

• •Run one peak-load test, one dependency-degradation test, and one failover test.
• •Verify idempotency for all retried writes and async consumers.
• •Track user-facing SLOs first: p95 latency, error rate, and successful throughput.

6) Trade-offs to Call Out in Interviews

• •Databases: SQL gives stronger transactional guarantees; NoSQL often gives better write scaling and flexibility.
• •Caching: Higher hit rate cuts latency/cost, but stale data and invalidation bugs become primary risks.
• •Microservices: Independent deployability improves scale but increases operational/debug complexity.
• •Search: Search index gives rich querying but introduces eventual consistency and index ops overhead.
• •Message Queues: Async pipelines absorb spikes well, but increase eventual-consistency complexity.

Practical Notes

• •Decompose into microservices: Catalog Service, Cart Service, Order Service, Payment Service, Inventory Service, Search Service, Recommendation Service, Fulfillment Service.
• •Cart: use a Dynamo-style AP system (always-writeable) - merge conflicts with 'union' strategy (no item is ever silently lost).
• •Inventory: two-phase approach - soft-reserve on 'add to cart', hard-reserve on checkout. Use pessimistic locking for the final decrement.

Why This Solution Works

Request path: The solution keeps ingress, service logic, and stateful dependencies separated so each layer can scale independently.

Reference flow: Web Clients -> Load Balancer -> API Gateway -> Core Service -> Primary NoSQL DB -> Redis Cache -> Message Queue -> Background Workers

Design strengths

• •Cache sits on the read path to absorb repeated queries and keep DB pressure stable.
• •Async queue/event bus isolates bursty workloads and supports retries without blocking synchronous requests.

Interview defense

• •This design makes bottlenecks explicit (ingress, core compute, persistence, async workers).
• •It supports progressive scaling without re-architecting the core request path.
• •It keeps correctness-sensitive state changes in durable systems while offloading background work asynchronously.