<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>benchmark on Mooncake</title>
    <link>https://mooncake.pgsty.com/tags/benchmark/</link>
    <description>Recent content in benchmark on Mooncake</description>
    <generator>Hugo</generator>
    <language>en-US</language>
    
    
    
      <lastBuildDate>Thu, 05 Nov 2026 19:17:16 +0800</lastBuildDate>
    
    
      <atom:link href="https://mooncake.pgsty.com/tags/benchmark/index.xml" rel="self" type="application/rss+xml" />
    
    <item>
        <title>How to improve DuckDB performance on Parquet</title>
        <link>https://mooncake.pgsty.com/blog/post/duckdb-parquet/</link>
        <pubDate>Wed, 22 Jan 2025 00:00:00 +0000</pubDate>
        
        <guid>https://mooncake.pgsty.com/blog/post/duckdb-parquet/</guid>
        <description>&lt;p&gt;Last week, we ran ClickBench on Postgres and ranked in the &lt;a href=&#34;https://mooncake.pgsty.com/blog/post/clickbench-v0.1/&#34;&gt;top 10&lt;/a&gt;. We showed that Postgres can be extended with &lt;a href=&#34;https://mooncake.pgsty.com/docs/intro/&#34;&gt;&lt;code&gt;pg_mooncake&lt;/code&gt;&lt;/a&gt; to support serious analytics workloads.&lt;/p&gt;&#xA;&lt;p&gt;In the announcement, we got a lot of questions on how &lt;code&gt;pg_mooncake&lt;/code&gt; outperforms DuckDB on Parquet files even though internally, it embeds DuckDB in Postgres and stores data in Parquet format.&lt;/p&gt;&#xA;&lt;blockquote&gt;&#xA;&lt;p&gt;See also the discussion on &lt;a href=&#34;https://x.com/mooncake_dev/status/1881844188905709880&#34;&gt;X / Twitter&lt;/a&gt;.&lt;/p&gt;&#xA;&lt;/blockquote&gt;&#xA;&lt;p&gt;The best part: &lt;code&gt;pg_mooncake&lt;/code&gt; does not modify DuckDB at all&amp;hellip;&lt;/p&gt;</description>
      </item>
    <item>
        <title>Clickbench says Postgres is a great analytics database</title>
        <link>https://mooncake.pgsty.com/blog/post/clickbench-v0.1/</link>
        <pubDate>Tue, 14 Jan 2025 00:00:00 +0000</pubDate>
        
        <guid>https://mooncake.pgsty.com/blog/post/clickbench-v0.1/</guid>
        <description>&lt;p&gt;&lt;strong&gt;TL;DR:&lt;/strong&gt;&#xA;We spent a few months optimizing PostgreSQL and made it to the &lt;a href=&#34;https://benchmark.clickhouse.com/&#34;&gt;&lt;strong&gt;Top 10 on ClickBench&lt;/strong&gt;&lt;/a&gt;, a benchmark typically dominated by specialized analytics databases.&lt;/p&gt;&#xA;&lt;p&gt;What&amp;rsquo;s more, all compute is within Postgres, and all tables are managed directly by PostgreSQL—it&amp;rsquo;s not a simple wrapper. This is the story of &lt;a href=&#34;https://mooncake.pgsty.com/docs/intro/&#34;&gt;&lt;code&gt;pg_mooncake&lt;/code&gt;&lt;/a&gt;.&lt;/p&gt;&#xA;&lt;img class=&#34;td-image&#34; src=&#34;https://mooncake.pgsty.com/images/blog/clickbench-blog.jpg&#34; alt=&#34;Clickbench&#34; loading=&#34;lazy&#34; decoding=&#34;async&#34;&gt;&lt;h2 id=&#34;what-is-clickbench&#34;&gt;What Is ClickBench?&#xA;&lt;/h2&gt;&#xA;&lt;p&gt;&lt;a href=&#34;https://github.com/ClickHouse/ClickBench/tree/main&#34;&gt;ClickBench&lt;/a&gt; is the definitive benchmark for real-time analytics databases, originally designed to showcase the performance of ClickHouse. It evaluates databases on their ability to handle real-world analytics workloads, including high-volume table scans and complex aggregations.&lt;/p&gt;</description>
      </item>
    
  </channel>
</rss>
