<?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>Tech on rkoush</title>
    <link>https://www.rkoush.com/categories/tech/</link>
    <description>Recent content in Tech on rkoush</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Fri, 01 May 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://www.rkoush.com/categories/tech/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>GPU architecture</title>
      <link>https://www.rkoush.com/posts/gpu/</link>
      <pubDate>Fri, 01 May 2026 00:00:00 +0000</pubDate>
      <guid>https://www.rkoush.com/posts/gpu/</guid>
      <description>&lt;h3 id=&#34;gpu&#34;&gt;GPU&lt;/h3&gt;
&lt;p&gt;GPU or graphics processing unit, are type of processors that are used to perform large floating point instructions. It gained intial popularity for it&amp;rsquo;s use in video processing or gaming and more recently for training large language models.&lt;/p&gt;
&lt;p&gt;But why are GPUs so fast and are they always faster than CPUs? The answer lies in the architecture. A GPU is designed
to perform single instruction over multiple data acronymed SIMD. A CPU usually contains Control Unit, Arithmatic Logic Unit, registers etc. More recently CPU have multiple cores, each cpu works like a mini cpu with it&amp;rsquo;s own CU and ALU. If a CPU has 4 cores, it can operate 4 instruction truly in parallel. GPU are different, they can 1000s of threads in parallel.&lt;/p&gt;</description>
    </item>
    <item>
      <title>TPU? Google&#39;s answer to Nvidia</title>
      <link>https://www.rkoush.com/posts/tpu/</link>
      <pubDate>Fri, 01 May 2026 00:00:00 +0000</pubDate>
      <guid>https://www.rkoush.com/posts/tpu/</guid>
      <description>&lt;h3 id=&#34;tensor-processing-units-tpu&#34;&gt;Tensor processing Units (TPU)&lt;/h3&gt;</description>
    </item>
  </channel>
</rss>
