<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Blog on Clearwater Engineering Blog</title><link>https://clearwater.cloud/blog/index.html</link><description>Recent content in Blog on Clearwater Engineering Blog</description><generator>Hugo</generator><language>en-AU</language><lastBuildDate>Fri, 20 Nov 2026 01:18:29 +0000</lastBuildDate><atom:link href="https://clearwater.cloud/blog/index.xml" rel="self" type="application/rss+xml"/><item><title>Kubernetes Right-Sizing Is an Operating Model, Not a Setting</title><link>https://clearwater.cloud/blog/kubernetes-right-sizing-operating-model.html</link><pubDate>Fri, 20 Nov 2026 01:18:29 +0000</pubDate><guid>https://clearwater.cloud/blog/kubernetes-right-sizing-operating-model.html</guid><description>&lt;p&gt;The AWS Containers blog recently published &lt;a href="https://aws.amazon.com/blogs/containers/how-salesforce-efficiently-right-sizes-kubernetes-workloads-at-scale-on-amazon-eks/"&gt;a detailed account of how Salesforce right-sizes Kubernetes workloads across its Amazon EKS fleet&lt;/a&gt;. It is worth reading if you run containers in production. The scale is extreme: more than 1,400 clusters and over $5 million in reported cumulative savings. But the useful lessons are not about scale. They are about how you change resource settings on live systems without breaking them.&lt;/p&gt;&#10;&lt;p&gt;The article&amp;rsquo;s opening problem will be familiar to anyone who has looked at a cloud bill: &amp;ldquo;Clusters running at low utilization consume compute that is paid for but never used.&amp;rdquo; Most Australian organisations we talk to are not running a thousand clusters. Many are running five to twenty, provisioned during a migration and never revisited. The waste is proportionally the same.&lt;/p&gt;</description></item><item><title>Your observability pipeline is a production system too</title><link>https://clearwater.cloud/blog/observability-pipeline-is-a-production-system.html</link><pubDate>Fri, 20 Nov 2026 01:04:08 +0000</pubDate><guid>https://clearwater.cloud/blog/observability-pipeline-is-a-production-system.html</guid><description>&lt;p&gt;Diana Todea&amp;rsquo;s recent piece on the CNCF blog, &lt;a href="https://www.cncf.io/blog/2026/11/17/is-your-opentelemetry-pipeline-becoming-a-performance-bottleneck/"&gt;Is your OpenTelemetry pipeline becoming a performance bottleneck?&lt;/a&gt;, walks through the main ways a Collector configuration can quietly burn resources. Her central warning is worth repeating to anyone who signs off on platform budgets: &amp;ldquo;In some cases, the infrastructure responsible for observing a system can become a performance problem of its own.&amp;rdquo;&lt;/p&gt;&#10;&lt;p&gt;We agree, and we&amp;rsquo;d go further. Most organisations treat observability as a tool they installed. In practice it is a distributed system they operate, with its own capacity limits, failure modes and running costs. It deserves the same design scrutiny as the services it watches.&lt;/p&gt;</description></item></channel></rss>