Stripe
Modeling Money Movement: Ledgers, State Machines, and Exactly-Once Effects
How Stripe models money movement with immutable ledger entries and explicit state machines to guarantee correctness even when networks and services fail.
3 min read
Tag
Distributed Systems
28 articles
Amazon
Why Amazon Bet Early on Queues: The Philosophy Behind SQS
How Amazon's early commitment to asynchronous, queue-based decoupling shaped SQS and became a default pattern across its service architecture.
3 min read
Netflix
Chaos Engineering at Netflix: Breaking Things on Purpose
How a painful database outage pushed Netflix to build Chaos Monkey, the Simian Army, and a culture that tests failure before failure finds you.
3 min read
Performance
Tail Latency: Why p99 Matters More Than Average
Why average response time hides the experience of your worst-served users, and how to measure, diagnose, and reduce tail latency in production systems.
3 min read
Uber
Cadence: Writing Workflow Code as If Failures Never Happen
How Uber's Cadence engine lets developers write long-running workflow logic as plain code while the platform handles retries and state durably.
3 min read
DoorDash
Reliability Engineering for a Three-Sided Marketplace
How DoorDash designs for graceful degradation and load shedding so a spike or partial outage on one side of its marketplace doesn't take down the rest.
3 min read
Twitter
Twitter's Timeline Problem: Push, Pull, and the Celebrity Edge Case
How Twitter combined fan-out-on-write and fan-out-on-read to keep timelines fast without letting celebrity accounts overwhelm the system.
3 min read
System Design
Backpressure in Streaming Systems
Bounded buffers, reactive pull-based demand, and log-based pull consumption: three ways streaming systems handle a slow consumer.
3 min read
Shopify
Background Jobs at Shopify Scale: Idempotency, Retries, and Interruptibility
How Shopify runs enormous volumes of background jobs safely, treating idempotency and graceful interruption as first-class design constraints.
3 min read
Meta
Haystack: Why Facebook Threw Out Filesystems for Photo Storage
How Facebook's Haystack replaced a filesystem-per-photo approach with a purpose-built object store to eliminate the metadata bottleneck of billions of images.
3 min read
System Design
Database Sharding Strategies That Survive Growth
Hash-based, range-based, and directory-based shard keys compared, and why resharding is the expensive part nobody plans for early.
2 min read
Amazon
The Dynamo Paper: How Amazon Learned to Love Eventual Consistency
How Amazon's shopping cart problem led to the 2007 Dynamo paper and reshaped how the industry thinks about availability versus consistency.
3 min read
Discord
Presence and Read States: Keeping Millions of Discord Clients in Sync
How Discord keeps online status and per-channel read state consistent across millions of simultaneously connected clients without collapsing under fan-out.
3 min read
Uber
Ringpop: Sharding Uber's Stateful Services Without a Central Coordinator
How Uber's Ringpop library used consistent hashing and a gossip protocol to shard stateful services without depending on a single coordinator.
3 min read
Twitter
Manhattan: Twitter's Multi-Tenant Distributed Database
How Twitter replaced a patchwork of Cassandra clusters with Manhattan, a purpose-built, multi-tenant distributed database with pluggable storage engines.
3 min read
LinkedIn
Espresso: The Document Store Behind LinkedIn's Member Data
Why LinkedIn built its own distributed document database to replace Oracle for online member data, and how it married timeline consistency with Kafka.
3 min read
Google
The MapReduce Paper: How Two Functions Started the Big Data Era
How Google's 2004 MapReduce paper turned a painful distributed-systems problem into a simple programming model — and accidentally created Hadoop.
3 min read
WhatsApp
Store-and-Forward: The Simple Idea Behind WhatsApp's Delivery Guarantees
How WhatsApp's store-and-forward architecture and layered acknowledgment model deliver messages reliably even when recipients are offline.
3 min read
System Design
Consistent Hashing, Explained From First Principles
How consistent hashing limits data movement when nodes join or leave, and why virtual nodes are what makes it actually balanced.
2 min read
Google
Spanner and TrueTime: How Google Made Globally Consistent Transactions Possible
How Google used GPS and atomic clocks to bound time uncertainty and give Spanner globally consistent ACID transactions across continents.
3 min read
DoorDash
Why DoorDash Moved Critical Transactions onto CockroachDB
DoorDash adopted CockroachDB to get horizontal scalability and strong consistency for order and payment data without the operational pain of manual sharding.
3 min read
Twitter
Snowflake: Twitter's Decentralized, Sortable ID Generator
How Twitter replaced a single MySQL sequence with a decentralized ID generator that packs time, machine, and sequence into a roughly sortable 64-bit ID.
3 min read
System Design
Idempotency in Distributed Systems: Making Retries Safe
Idempotency keys, atomic check-and-write, and why PUT being idempotent by spec doesn't make it the right tool for a retried write.
2 min read
Meta
Scaling Memcached: How Facebook Made Look-Aside Caching Work at Web Scale
How Facebook turned a simple key-value cache into a cluster of thousands of memcached servers without drowning in stale reads and thundering herds.
3 min read