<?xml version="1.0" encoding="UTF-8" ?>
<rss version="2.0">
    <channel>
      <title>Raghav&#039;s Wiki</title>
      <link>https://wiki.raghavrastogi.xyz</link>
      <description>Last 10 notes on Raghav&#039;s Wiki</description>
      <generator>Quartz -- quartz.jzhao.xyz</generator>
      <item>
    <title>Connection Pooling</title>
    <link>https://wiki.raghavrastogi.xyz/concepts/connection-pooling</link>
    <guid>https://wiki.raghavrastogi.xyz/concepts/connection-pooling</guid>
    <description><![CDATA[ Definition A technique where a proxy layer maintains a pool of open database connections that are reused across many application clients — avoiding the overhead of establishing a new connection per request. ]]></description>
    <pubDate>Sat, 25 Apr 2026 09:19:37 GMT</pubDate>
  </item><item>
    <title>Database Replication</title>
    <link>https://wiki.raghavrastogi.xyz/concepts/database-replication</link>
    <guid>https://wiki.raghavrastogi.xyz/concepts/database-replication</guid>
    <description><![CDATA[ Definition The process of copying data from a primary database instance to one or more replicas, enabling read scaling, fault tolerance, and geographic distribution. ]]></description>
    <pubDate>Sat, 25 Apr 2026 09:19:37 GMT</pubDate>
  </item><item>
    <title>Load Shedding</title>
    <link>https://wiki.raghavrastogi.xyz/concepts/load-shedding</link>
    <guid>https://wiki.raghavrastogi.xyz/concepts/load-shedding</guid>
    <description><![CDATA[ Definition The deliberate rejection of incoming requests to protect a system from overload — accepting reduced throughput now to maintain stability and preserve capacity for the most important work. ]]></description>
    <pubDate>Sat, 25 Apr 2026 09:19:37 GMT</pubDate>
  </item><item>
    <title>PostgreSQL Scaling</title>
    <link>https://wiki.raghavrastogi.xyz/concepts/postgresql-scaling</link>
    <guid>https://wiki.raghavrastogi.xyz/concepts/postgresql-scaling</guid>
    <description><![CDATA[ Definition The set of techniques used to scale PostgreSQL beyond a single instance — primarily by scaling reads horizontally via replicas and routing write-heavy workloads to sharded alternatives, rather than sharding PostgreSQL itself. ]]></description>
    <pubDate>Sat, 25 Apr 2026 09:19:37 GMT</pubDate>
  </item><item>
    <title>Thundering Herd</title>
    <link>https://wiki.raghavrastogi.xyz/concepts/thundering-herd</link>
    <guid>https://wiki.raghavrastogi.xyz/concepts/thundering-herd</guid>
    <description><![CDATA[ Definition A failure pattern where many clients simultaneously request the same resource — typically after a cache miss or service recovery — causing a sudden spike that overwhelms the backend. ]]></description>
    <pubDate>Sat, 25 Apr 2026 09:19:37 GMT</pubDate>
  </item><item>
    <title>Bohan Zhang</title>
    <link>https://wiki.raghavrastogi.xyz/entities/bohan-zhang</link>
    <guid>https://wiki.raghavrastogi.xyz/entities/bohan-zhang</guid>
    <description><![CDATA[ About Bohan Zhang is a Member of Technical Staff at OpenAI, working on production database infrastructure. ]]></description>
    <pubDate>Sat, 25 Apr 2026 09:19:37 GMT</pubDate>
  </item><item>
    <title>OpenAI Blog</title>
    <link>https://wiki.raghavrastogi.xyz/entities/openai-blog</link>
    <guid>https://wiki.raghavrastogi.xyz/entities/openai-blog</guid>
    <description><![CDATA[ About OpenAI’s official engineering and research blog. Covers model releases, infrastructure engineering, safety research, and policy. ]]></description>
    <pubDate>Sat, 25 Apr 2026 09:19:37 GMT</pubDate>
  </item><item>
    <title>Scaling PostgreSQL to power 800 million ChatGPT users</title>
    <link>https://wiki.raghavrastogi.xyz/sources/scaling-postgresql-openai</link>
    <guid>https://wiki.raghavrastogi.xyz/sources/scaling-postgresql-openai</guid>
    <description><![CDATA[ Summary A first-person engineering account by Bohan Zhang of how OpenAI scaled PostgreSQL to millions of queries per second for 800 million ChatGPT and API users — without sharding PostgreSQL itself. ]]></description>
    <pubDate>Sat, 25 Apr 2026 09:19:37 GMT</pubDate>
  </item><item>
    <title>How Uber Conquered Database Overload: The Journey from Static Rate-Limiting to Intelligent Load Management</title>
    <link>https://wiki.raghavrastogi.xyz/sources/uber-intelligent-load-management</link>
    <guid>https://wiki.raghavrastogi.xyz/sources/uber-intelligent-load-management</guid>
    <description><![CDATA[ Summary A multi-author engineering post from Uber’s Storage Platform team tracing the full evolution of overload protection for Docstore and Schemaless — Uber’s in-house distributed databases built on MySQL, serving tens of millions of QPS for 170M monthly active users. ]]></description>
    <pubDate>Sat, 25 Apr 2026 09:19:37 GMT</pubDate>
  </item><item>
    <title>Uber Engineering Blog</title>
    <link>https://wiki.raghavrastogi.xyz/entities/uber-engineering-blog</link>
    <guid>https://wiki.raghavrastogi.xyz/entities/uber-engineering-blog</guid>
    <description><![CDATA[ About Uber’s engineering blog covering infrastructure, distributed systems, databases, ML, and platform engineering. ]]></description>
    <pubDate>Sat, 25 Apr 2026 08:08:25 GMT</pubDate>
  </item>
    </channel>
  </rss>