<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Ebs on nanta - Data Engineering</title><link>https://nanta-data.dev/en/tags/ebs/</link><description>Recent content in Ebs on nanta - Data Engineering</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>© 2026 nanta</copyright><lastBuildDate>Fri, 27 Feb 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://nanta-data.dev/en/tags/ebs/index.xml" rel="self" type="application/rss+xml"/><item><title>Trino Alluxio Cache PoC: EBS Throughput Was the Bottleneck</title><link>https://nanta-data.dev/en/posts/trino-alluxio-cache-poc/</link><pubDate>Fri, 27 Feb 2026 00:00:00 +0000</pubDate><guid>https://nanta-data.dev/en/posts/trino-alluxio-cache-poc/</guid><description>We ran a PoC of Trino&amp;rsquo;s Alluxio-based file system cache in a production OLAP environment. Adding the cache alone didn&amp;rsquo;t help much. The default EBS throughput of 125 MiB/s was the bottleneck. After bumping it to 1000 MiB/s, query performance improved visibly and S3 API costs dropped by ~$3,200/month.</description></item></channel></rss>