<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Storage on AI VOID</title><link>https://ai-blog.noorshomelab.dev/tags/storage/</link><description>Recent content in Storage on AI VOID</description><generator>Hugo</generator><language>en</language><lastBuildDate>Mon, 20 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://ai-blog.noorshomelab.dev/tags/storage/index.xml" rel="self" type="application/rss+xml"/><item><title>Lifecycle Management: State, Storage, and I/O</title><link>https://ai-blog.noorshomelab.dev/smolvm-architecture-2026-04/lifecycle-management-state-storage-io/</link><pubDate>Mon, 20 Apr 2026 00:00:00 +0000</pubDate><guid>https://ai-blog.noorshomelab.dev/smolvm-architecture-2026-04/lifecycle-management-state-storage-io/</guid><description>&lt;p&gt;Managing the lifecycle of a virtual machine—from its initial setup to saving and restoring its exact state—is a core challenge in virtualization. For platforms like &lt;code&gt;smolvm&lt;/code&gt;, this isn&amp;rsquo;t just about basic operations; it&amp;rsquo;s about redefining expectations with sub-second cold starts and highly portable, stateful environments.&lt;/p&gt;
&lt;p&gt;This chapter will guide you through the intricate architectural decisions that enable &lt;code&gt;smolvm&lt;/code&gt; to deliver on these promises. We&amp;rsquo;ll dissect how it handles VM state, optimizes storage, and orchestrates I/O across diverse operating systems. Understanding these internal mechanisms is vital for any developer or architect aiming to leverage &lt;code&gt;smolvm&lt;/code&gt; for rapid development, consistent testing, or streamlined software distribution.&lt;/p&gt;</description></item><item><title>Optimizing ML Tensor Storage and Transfer</title><link>https://ai-blog.noorshomelab.dev/openzl-mastery-2026/project-optimizing-ml-tensor-storage/</link><pubDate>Mon, 26 Jan 2026 00:00:00 +0000</pubDate><guid>https://ai-blog.noorshomelab.dev/openzl-mastery-2026/project-optimizing-ml-tensor-storage/</guid><description>&lt;h2 id="optimizing-ml-tensor-storage-and-transfer"&gt;Optimizing ML Tensor Storage and Transfer&lt;/h2&gt;
&lt;p&gt;Welcome back, future data compression wizard! In this chapter, we&amp;rsquo;re diving into one of the most exciting and impactful applications of OpenZL: &lt;strong&gt;optimizing the storage and transfer of Machine Learning (ML) tensors.&lt;/strong&gt; If you&amp;rsquo;ve ever worked with large ML models, you know that tensors – the multi-dimensional arrays that represent everything from model weights to activation maps – can become incredibly bulky. This bulk leads to slow loading times, high storage costs, and bottlenecks in data transfer, especially in distributed training or inference scenarios.&lt;/p&gt;</description></item></channel></rss>