<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>AI Architecture Notes</title><description>AI explains system design and architecture concepts in simple terms. No fluff, no jargon walls — just clear explanations with diagrams.</description><link>https://www.aiarchitecturenotes.in/</link><item><title>Redis Explained</title><link>https://www.aiarchitecturenotes.in/articles/redis/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/redis/</guid><description>How does a single thread handle 100K ops/sec? Dive into the event loop, memory layout, and why Instagram chose Redis for 300M mappings.</description><pubDate>Sun, 01 Mar 2026 00:00:00 GMT</pubDate></item><item><title>Kafka Explained</title><link>https://www.aiarchitecturenotes.in/articles/kafka/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/kafka/</guid><description>Why is sequential disk I/O faster than random memory access? Zero-copy transfers, partition replication, and how LinkedIn handles 7 trillion messages/day.</description><pubDate>Mon, 02 Mar 2026 00:00:00 GMT</pubDate></item><item><title>Load Balancers Explained</title><link>https://www.aiarchitecturenotes.in/articles/load-balancers/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/load-balancers/</guid><description>What happens in the 1ms an ALB adds to your request? L4 vs L7, consistent hashing math, and how Cloudflare routes 50M req/sec.</description><pubDate>Tue, 03 Mar 2026 00:00:00 GMT</pubDate></item><item><title>Database Sharding Explained</title><link>https://www.aiarchitecturenotes.in/articles/database-sharding/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/database-sharding/</guid><description>The $24K/month server that still isn&apos;t enough. How Instagram shards PostgreSQL, Discord handles trillions of messages, and when NOT to shard.</description><pubDate>Wed, 04 Mar 2026 00:00:00 GMT</pubDate></item><item><title>Rate Limiting Explained</title><link>https://www.aiarchitecturenotes.in/articles/rate-limiting/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/rate-limiting/</guid><description>Why one Elon tweet can take down your API. Token buckets, Cloudflare&apos;s 99.997% accurate algorithm, and distributed rate limiting across 300 data centers.</description><pubDate>Thu, 05 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Did Justin Bieber Break Instagram?</title><link>https://www.aiarchitecturenotes.in/articles/justin-bieber-problem/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/justin-bieber-problem/</guid><description>The thundering herd problem explained through celebrity posts, cache stampedes, and the solutions that actually work.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>Why Can&apos;t Distributed Systems Have It All?</title><link>https://www.aiarchitecturenotes.in/articles/cap-theorem/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/cap-theorem/</guid><description>The CAP theorem says pick two: consistency, availability, partition tolerance. Google Spanner, DynamoDB, and the real tradeoffs.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>Why Does Every Database Have a Leader?</title><link>https://www.aiarchitecturenotes.in/articles/leader-follower/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/leader-follower/</guid><description>Leader-follower replication powers PostgreSQL, MySQL, MongoDB, and Redis. One node writes, the rest copy.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>What Happens When Both Databases Accept Writes?</title><link>https://www.aiarchitecturenotes.in/articles/active-replication/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/active-replication/</guid><description>Active-active replication lets every node accept writes. How CockroachDB and DynamoDB Global Tables handle it.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Does a Database Survive a Crash?</title><link>https://www.aiarchitecturenotes.in/articles/wal/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/wal/</guid><description>Write-Ahead Logging is the reason your data survives power failures, kernel panics, and OOM kills.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Do Distributed Systems Agree on Anything?</title><link>https://www.aiarchitecturenotes.in/articles/consensus/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/consensus/</guid><description>Paxos, Raft, and the consensus problem. How etcd, ZooKeeper, and CockroachDB get multiple nodes to agree.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Do You Track Causality Across Servers?</title><link>https://www.aiarchitecturenotes.in/articles/vector-clocks/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/vector-clocks/</guid><description>Wall clocks lie in distributed systems. Vector clocks track causal ordering without synchronized time.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Do 1,000 Servers Agree Without a Leader?</title><link>https://www.aiarchitecturenotes.in/articles/gossip-protocol/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/gossip-protocol/</guid><description>Gossip protocols spread information like rumors in a crowd. Cassandra, DynamoDB, and Consul use them.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Does Figma Let 50 People Edit at Once?</title><link>https://www.aiarchitecturenotes.in/articles/crdts/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/crdts/</guid><description>CRDTs let multiple users edit the same data without coordination. No locks, no conflicts, mathematically guaranteed.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Do Microservices Handle Transactions?</title><link>https://www.aiarchitecturenotes.in/articles/saga-pattern/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/saga-pattern/</guid><description>The Saga pattern breaks distributed transactions into compensatable steps. How Uber, Netflix, and Airbnb do it.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Does DynamoDB Distribute Data Across Nodes?</title><link>https://www.aiarchitecturenotes.in/articles/consistent-hashing/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/consistent-hashing/</guid><description>Consistent hashing lets distributed systems add and remove nodes without reshuffling all the data.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Did 43 Seconds Break GitHub for 24 Hours?</title><link>https://www.aiarchitecturenotes.in/articles/github-outage/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/github-outage/</guid><description>A routine network maintenance caused a 43-second partition, MySQL failover triggered, and GitHub spent 24 hours recovering.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Does Discord Store Trillions of Messages?</title><link>https://www.aiarchitecturenotes.in/articles/discord-messages/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/discord-messages/</guid><description>Discord migrated from MongoDB to Cassandra to ScyllaDB. Each migration solved one problem and created another.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Did a Single Regex Take Down 15% of the Internet?</title><link>https://www.aiarchitecturenotes.in/articles/cloudflare-outage/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/cloudflare-outage/</guid><description>A regular expression in Cloudflare&apos;s WAF caused every CPU core across their global network to spike to 100%.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>Why Does Your p99 Latency Ruin Everything?</title><link>https://www.aiarchitecturenotes.in/articles/tail-latency/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/tail-latency/</guid><description>Your average latency is 5ms. Your p99 is 500ms. At Google&apos;s scale, every user hits tail latency.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Does MQTT Keep Billions of Devices Talking?</title><link>https://www.aiarchitecturenotes.in/articles/mqtt/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/mqtt/</guid><description>How a protocol designed for oil pipelines in 1999 now powers the entire IoT world.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>What Happens in the 1.5 Round Trips Before Your Data Flows?</title><link>https://www.aiarchitecturenotes.in/articles/tcp-handshake/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/tcp-handshake/</guid><description>Every TCP connection starts with a three-way handshake. SYN, SYN-ACK, ACK.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>What Happens Between Typing a URL and the First Byte?</title><link>https://www.aiarchitecturenotes.in/articles/dns-resolution/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/dns-resolution/</guid><description>DNS resolves the name to an IP through root servers, TLD servers, and authoritative nameservers.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>How Does Your Chat App Get Messages Instantly?</title><link>https://www.aiarchitecturenotes.in/articles/websockets/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/websockets/</guid><description>WebSockets upgrade an HTTP connection into a persistent, full-duplex channel.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item><item><title>Why Does Opening a Database Connection Take 30ms?</title><link>https://www.aiarchitecturenotes.in/articles/connection-pooling/</link><guid isPermaLink="true">https://www.aiarchitecturenotes.in/articles/connection-pooling/</guid><description>Connection pooling reuses connections to avoid TCP handshake, TLS, and auth costs.</description><pubDate>Mon, 09 Mar 2026 00:00:00 GMT</pubDate></item></channel></rss>