System Design

38 topics across 8 categories · 0/38 read

Common Questions

⊣ URL ShortenerDesign a service like bit.ly. Covers hashing strategies, base62 encoding, collision handling, redirect performance, and analytics tracking.

⊣ Chat SystemDesign a real-time messaging system like WhatsApp or Slack. Covers WebSocket connections, message ordering, presence, and offline message delivery.

⊣ News Feed / TimelineDesign a social media feed like Twitter or Instagram. Covers fan-out-on-write vs fan-out-on-read, ranking, and real-time updates.

⊣ Rate LimitingProtect services from abuse with token bucket, leaky bucket, fixed window, and sliding window algorithms. Distributed rate limiting with Redis.

⊣ Notification SystemDesign a multi-channel notification system supporting push, SMS, and email. Covers priority queues, rate limiting, user preferences, and delivery guarantees.

⊣ Web CrawlerDesign a web crawler like Googlebot. Covers URL frontier, politeness policies, distributed crawling, deduplication, and handling traps.

⊣ Search AutocompleteDesign a typeahead system like Google search suggestions. Covers trie data structures, top-K queries, prefix caching, and real-time ranking.

⊣ File Storage ServiceDesign Google Drive or Dropbox. Covers file chunking, sync conflicts, metadata vs blob storage, versioning, and real-time collaboration.

⊣ Video Streaming PlatformDesign YouTube or Netflix. Covers video transcoding, adaptive bitrate streaming, CDN delivery, view counting at scale, and recommendations.

⊣ Ride-Sharing ServiceDesign Uber or Lyft. Covers geospatial indexing, real-time driver matching, ETA calculation, surge pricing, and location tracking at scale.

⊣ Distributed Key-Value StoreDesign a system like DynamoDB or Cassandra. Covers consistent hashing, vector clocks, quorum reads/writes, and conflict resolution.

⊣ Search EngineDesign a web search engine. Covers inverted indexes, TF-IDF ranking, PageRank, query parsing, index sharding, and serving optimization.

⊣ Payment SystemDesign a payment system like Stripe. Covers idempotency, double-entry bookkeeping, payment state machines, webhooks, and reconciliation.

⊣ Load BalancingDistribute traffic across servers using round-robin, least connections, consistent hashing, and weighted algorithms. L4 vs L7 load balancing.

Fundamentals

⊣ Scaling: Vertical vs HorizontalUnderstand when to scale up a single machine versus scaling out across many. Covers the trade-offs, bottlenecks, and when each approach makes sense.

⊣ CAP TheoremYou can only guarantee two of three: consistency, availability, and partition tolerance. Learn how real systems make this trade-off and what it means for your design.

⊣ Consistency ModelsStrong, eventual, causal, and read-your-writes consistency. Understand the spectrum from linearizability to eventual consistency and when each is appropriate.

⊣ Latency and ThroughputThe two fundamental performance metrics. Learn how to reason about latency budgets, throughput bottlenecks, and the relationship between them.

Networking & Communication

⊣ REST API DesignDesign clean, predictable HTTP APIs. Covers resource naming, HTTP methods, status codes, pagination, versioning, and common anti-patterns.

⊣ WebSockets and Long PollingReal-time communication patterns for chat, notifications, and live updates. Compare WebSockets, SSE, and long polling trade-offs.

⊣ gRPC and Protocol BuffersHigh-performance RPC framework using binary serialization. When to choose gRPC over REST, and how to design service contracts with Protocol Buffers.

⊣ DNS and CDNsHow DNS resolution works, DNS-based load balancing, and how CDNs cache content at the edge to reduce latency for global users.

Data Storage

⊣ SQL vs NoSQLRelational databases vs document stores, key-value stores, wide-column, and graph databases. How to choose the right storage model for your use case.

⊣ Database ShardingHorizontally partition data across multiple database instances. Covers hash-based, range-based, and directory-based sharding strategies and their trade-offs.

⊣ Replication and FailoverLeader-follower, multi-leader, and leaderless replication. How to achieve high availability, handle failover, and manage replication lag.

⊣ Database IndexingB-trees, LSM trees, hash indexes, and composite indexes. Understand how indexes speed up reads, slow down writes, and when to use each type.

Caching

⊣ Caching StrategiesCache-aside, write-through, write-behind, and read-through patterns. When to cache, what to cache, and how to handle cache invalidation.

⊣ Distributed Caching with RedisUsing Redis as a distributed cache, session store, and rate limiter. Covers eviction policies, persistence options, and cluster mode.

⊣ CDN and Edge CachingPush content to edge locations close to users. Covers cache-control headers, invalidation strategies, and when to use a CDN vs application-level cache.

Scalability Patterns

⊣ Load BalancingDistribute traffic across servers using round-robin, least connections, consistent hashing, and weighted algorithms. L4 vs L7 load balancing.

⊣ Rate LimitingProtect services from abuse with token bucket, leaky bucket, fixed window, and sliding window algorithms. Distributed rate limiting with Redis.

⊣ Circuit Breaker PatternPrevent cascading failures by detecting when a downstream service is unhealthy and failing fast. Covers closed, open, and half-open states.

⊣ API GatewayA single entry point for all client requests. Handles authentication, rate limiting, routing, and protocol translation. The front door of microservices.

Messaging & Async

⊣ Message QueuesDecouple producers from consumers using queues. Covers RabbitMQ, SQS, and when to use point-to-point queues vs publish-subscribe topics.

⊣ Event-Driven ArchitectureBuild loosely coupled systems using events. Covers event sourcing, CQRS, and the difference between event notification and event-carried state transfer.

⊣ Kafka and Event StreamingDistributed commit log for high-throughput event streaming. Covers partitions, consumer groups, exactly-once semantics, and when Kafka is the right choice.

Classic Designs

⊣ URL ShortenerDesign a service like bit.ly. Covers hashing strategies, base62 encoding, collision handling, redirect performance, and analytics tracking.

⊣ Chat SystemDesign a real-time messaging system like WhatsApp or Slack. Covers WebSocket connections, message ordering, presence, and offline message delivery.

⊣ News Feed / TimelineDesign a social media feed like Twitter or Instagram. Covers fan-out-on-write vs fan-out-on-read, ranking, and real-time updates.

⊣ Notification SystemDesign a multi-channel notification system supporting push, SMS, and email. Covers priority queues, rate limiting, user preferences, and delivery guarantees.

⊣ Distributed Key-Value StoreDesign a system like DynamoDB or Cassandra. Covers consistent hashing, vector clocks, quorum reads/writes, and conflict resolution.

⊣ Web CrawlerDesign a web crawler like Googlebot. Covers URL frontier, politeness policies, distributed crawling, deduplication, and handling traps.

⊣ Search AutocompleteDesign a typeahead system like Google search suggestions. Covers trie data structures, top-K queries, prefix caching, and real-time ranking.

⊣ File Storage ServiceDesign Google Drive or Dropbox. Covers file chunking, sync conflicts, metadata vs blob storage, versioning, and real-time collaboration.

⊣ Video Streaming PlatformDesign YouTube or Netflix. Covers video transcoding, adaptive bitrate streaming, CDN delivery, view counting at scale, and recommendations.

⊣ Ride-Sharing ServiceDesign Uber or Lyft. Covers geospatial indexing, real-time driver matching, ETA calculation, surge pricing, and location tracking at scale.

⊣ Search EngineDesign a web search engine. Covers inverted indexes, TF-IDF ranking, PageRank, query parsing, index sharding, and serving optimization.

⊣ Payment SystemDesign a payment system like Stripe. Covers idempotency, double-entry bookkeeping, payment state machines, webhooks, and reconciliation.

Infrastructure

⊣ Distributed ConsensusHow nodes agree on values in an unreliable network. Covers Paxos, Raft, and ZAB protocols, and why consensus is the backbone of distributed systems.

⊣ Consistent HashingDistribute data across nodes while minimizing redistribution when nodes join or leave. The foundation of distributed caches and databases.

⊣ Bloom FiltersSpace-efficient probabilistic data structure for membership testing. Covers false positive rates, sizing, and real-world uses in databases and caches.

⊣ Monitoring and ObservabilityLogs, metrics, and traces — the three pillars. Design monitoring infrastructure that helps you detect, diagnose, and resolve issues in production.