<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Prefix on StorageNews</title>
    <link>https://storagenews.top/tags/prefix/</link>
    <description>Recent content in Prefix on StorageNews</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Fri, 24 Jul 2026 12:24:14 +0000</lastBuildDate>
    <atom:link href="https://storagenews.top/tags/prefix/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>S3 Prefixes Unlock 5,500 Requests Per Second</title>
      <link>https://storagenews.top/posts/s3-prefixes-hit-5500-requestssecond/</link>
      <pubDate>Tue, 14 Jul 2026 00:00:00 +0000</pubDate>
      <author>StorageNews</author>
      <guid>https://storagenews.top/posts/s3-prefixes-hit-5500-requestssecond/</guid>
      <description>Learn how partitioned prefixes sustain 5,500 requests per second while byterange fetches eliminate bottlenecks for large AI datasets.</description>
    </item>
    <item>
      <title>Prefix caching cuts LLM latency by 70%</title>
      <link>https://storagenews.top/posts/gke-inference-gateway-prefix-caching-dont-sign-up-for-the-cache-tax/</link>
      <pubDate>Thu, 11 Jun 2026 00:00:00 +0000</pubDate>
      <author>StorageNews</author>
      <guid>https://storagenews.top/posts/gke-inference-gateway-prefix-caching-dont-sign-up-for-the-cache-tax/</guid>
      <description>GKE Inference Gateway uses prefix caching to cut time-to-first-token latency by over 70%, eliminating redundant computation in AI pipelines.</description>
    </item>
  </channel>
</rss>
