Design YouTube

Interview-Ready Approach

1) Clarify Scope and SLOs

• •Problem statement: Design YouTube's video platform - upload, transcode, adaptive streaming, recommendations, and comments at planetary scale.
• •Design for a peak load target around 80,000 RPS (including burst headroom).
• •Monthly active users: 2,500,000,000
• •Hours of video watched/day: 1,000,000,000
• •Video uploads/minute: ~500 hours
• •Transcoding variants per video: 20+
• •Total video storage: Exabytes

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

• •CDN: Serve static and cacheable content from edge and keep origin strictly for misses and dynamic requests.
• •Media Processing: Split ingest, transform, and delivery into independent stages with async orchestration.
• •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.
• •Storage: Use object storage for large blobs and keep metadata/authorization separate in the API tier.
• •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.

4) Reliability and Failure Strategy

• •Define cache keys and purge workflows before launch to avoid stale/global outages.
• •Store original media durably and make transforms replayable.
• •Use strong write constraints (transactions or conditional writes) and explicit backup/restore strategy.
• •Bound staleness with TTL + invalidation hooks for critical entities.
• •Enforce lifecycle policies, retention tiers, and checksum validation.

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

• •CDN: Long TTL improves latency/cost; short TTL improves freshness.
• •Media Processing: Pre-processing improves playback UX, but requires substantial compute/storage budget.
• •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.
• •Storage: Object storage is cheap and durable, but random low-latency reads are weaker than databases/caches.

Practical Notes

• •Split video into segments and transcode in parallel across a GPU worker fleet - each segment is independent.
• •Hot/cold storage tiering: top 10% of videos (by views) on SSD at edge; long-tail on HDD/tape at origin.
• •Recommendation: two-stage pipeline - candidate generation (fast, broad) → ranking (ML model, precise). Pre-compute home feed offline, refresh on-demand.

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 -> DNS -> CDN Edge -> Load Balancer -> API Gateway -> Core Service -> Primary SQL DB -> Redis Cache

Design strengths

• •Cache sits on the read path to absorb repeated queries and keep DB pressure stable.
• •Media processing is handled by background workers so user-facing latency stays low.

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.