<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>hashing on chrismoos</title>
    <link>/tags/hashing/</link>
    <description>Recent content in hashing on chrismoos</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en-us</language>
    <copyright>&amp;copy; 2020 Chris Moos</copyright>
    <lastBuildDate>Sat, 04 Jun 2011 16:49:11 +0000</lastBuildDate><atom:link href="/tags/hashing/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>libhashring - high performance consistent hashing</title>
      <link>/2011/06/04/libhashring-high-performance-consistent-hashing/</link>
      <pubDate>Sat, 04 Jun 2011 16:49:11 +0000</pubDate>
      
      <guid>/2011/06/04/libhashring-high-performance-consistent-hashing/</guid>
      <description>In a previous post I talked about the need to shard Redis data and how I accomplished this by adding shard/hashing support to erldis, an Erlang client for Redis. This solution proved to work well, it distributed our data very well amongst many Redis servers &amp;ndash; but there was one problem. Performance.
In the change I made to erldis, the hash ring was stored in ETS (an erlang memory store) and anytime a key was hashed, the ring had to be retrieved from ETS.</description>
    </item>
    
  </channel>
</rss>
