System Design — Page 16 of 21
Browsing page 16 of 21 — 243 articles on system design.
Designing a Graph Database: Storage Engines, Query Optimization, and Traversal Patterns for Connected Data
Graph databases store connected data differently from relational systems. This guide covers property graph vs RDF models, adjacency list storage, index-free adjacency, multi-hop query optimization, sharding strategies, and traversal algorithms with TypeScript examples.
Designing a Session Management System: Tokens, Storage Backends, and Revocation at Scale
A deep-dive into session token design, storage backend tradeoffs, revocation patterns, and multi-region consistency for production session management systems. Covers opaque vs JWT tokens, Redis vs DynamoDB vs database-backed storage, individual and bulk revocation, session fixation, and scaling strategies.
Designing a Change Data Capture Pipeline: Log-Based CDC, Debezium, and Real-Time Data Synchronization
Polling-based sync breaks under load. Learn how log-based CDC works using PostgreSQL WAL and MySQL binlog, how Debezium fits into the architecture, and how to build a production-grade CDC pipeline with TypeScript consumers, schema evolution handling, and ordering guarantees.
Designing a Service Discovery System: Client-Side, Server-Side, and DNS-Based Patterns for Microservices
A deep dive into the three main service discovery patterns for microservices: client-side with a registry, server-side with a load balancer, and DNS-based. Covers health checking, consistency tradeoffs, and TypeScript implementations for health-aware routing.
Write-Ahead Logs: How Databases and Distributed Systems Guarantee Durability
A deep-dive into the write-ahead log pattern that underpins crash recovery, replication, and point-in-time restore in PostgreSQL, SQLite, Kafka, etcd, and beyond. Covers WAL structure, checkpointing, log compaction, fsync cost, group commit, and production monitoring.
Designing a Distributed ID Generator: UUIDs, Snowflake IDs, and Coordination-Free Sequences at Scale
A deep dive into generating unique identifiers in distributed systems. Covers UUID v4 and v7, Twitter's Snowflake approach, ULIDs, and coordination-free strategies, with TypeScript implementations, tradeoff comparisons, and production considerations for clock skew and multi-region deployments.
Designing a Distributed Key-Value Store: Partitioning, Replication, and Consistency Guarantees
A deep dive into distributed key-value store design covering consistent hashing, virtual nodes, replication strategies, consistency models, conflict resolution with vector clocks and CRDTs, and production considerations like compaction and read repair.
Designing a Real-Time Fraud Detection System: Rules Engines, ML Scoring, and Stream Processing at Scale
How to architect a real-time fraud detection pipeline using a three-layer approach: deterministic rule engines, ML model scoring, and stream processing for event correlation. Covers data modeling, feature stores, latency budgets, and the false positive tradeoff.
Designing a Search Engine: Inverted Indexes, Relevance Scoring, and Query Processing at Scale
A deep-dive into search engine internals: inverted index construction, tokenization pipelines, BM25 scoring, query parsing and execution, index sharding and replication, near-real-time indexing, faceted search, and the precision vs. recall tradeoffs that define production search systems. TypeScript examples throughout.
Designing a Data Pipeline: ETL vs ELT, Batch vs Streaming, and Orchestration Patterns for Production
A practical comparison of ETL vs ELT, batch vs streaming, and orchestration tools for production data pipelines. Covers schema evolution, backfill strategies, exactly-once semantics, and monitoring with TypeScript code examples.
Designing a Geospatial Search System: Spatial Indexes, Proximity Queries, and Location-Based Services at Scale
A deep-dive into spatial indexing strategies, proximity query internals, and the architecture behind location-based services that handle millions of location updates per second.
Designing a Recommendation Engine: Collaborative Filtering, Embeddings, and Real-Time Personalization
A deep dive into recommendation system architecture: collaborative filtering approaches, content-based methods, embedding-based retrieval, the cold start problem, real-time vs batch scoring, A/B testing quality, and the infrastructure required to serve recommendations at low latency in production.