<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Series-Lakehouse-in-Production on Sakura Sky: Cloud, Data, Security</title><link>https://www.sakurasky.com/tags/series-lakehouse-in-production/</link><description>Recent content in Series-Lakehouse-in-Production on Sakura Sky: Cloud, Data, Security</description><generator>Hugo</generator><language>en-US</language><copyright>Sakura Sky</copyright><lastBuildDate>Tue, 13 Oct 2026 00:00:00 -0400</lastBuildDate><atom:link href="https://www.sakurasky.com/tags/series-lakehouse-in-production/index.xml" rel="self" type="application/rss+xml"/><item><title>Keeping Raw Data You May Be Asked to Delete</title><link>https://www.sakurasky.com/blog/lakehouse-bronze-data-erasure-part-1/</link><pubDate>Tue, 13 Oct 2026 00:00:00 -0400</pubDate><guid>https://www.sakurasky.com/blog/lakehouse-bronze-data-erasure-part-1/</guid><description>&lt;style&gt;&#10; &#10;.series-pull-quote {&#10; float: right;&#10; clear: both;&#10; width: 340px;&#10; margin-left: 2rem;&#10; margin-bottom: 1.5rem;&#10; border: 1px solid #e9ecef;&#10; border-radius: 6px;&#10; overflow: hidden;&#10; background-color: #ffffff;&#10; &#10; position: relative;&#10; z-index: 20;&#10;}&#10;&#10; &#10;@media (max-width: 768px) {&#10; .series-pull-quote {&#10; float: none;&#10; clear: both;&#10; width: 100%;&#10; margin-left: 0;&#10; }&#10;}&#10;&#10;.series-header {&#10; background-color: #f8f9fa;&#10; color: #6c757d;&#10; font-size: 0.65rem;&#10; font-weight: 700;&#10; letter-spacing: 0.05em;&#10; padding: 0.8rem 1.25rem;&#10; border-bottom: 1px solid #e9ecef;&#10;}&#10;&#10;.part-label {&#10; display: block;&#10; font-size: 0.6rem;&#10; color: #adb5bd;&#10; margin-bottom: 0.15rem;&#10; text-transform: uppercase;&#10;}&#10;&#10;.nav-link-text {&#10; font-size: 0.85rem;&#10; line-height: 1.4;&#10; color: #495057;&#10; text-decoration: none;&#10; &#10; position: relative;&#10; z-index: 21;&#10;}&#10;&#10; &#10;.active-part {&#10; border-left: 4px solid #000 !important;&#10; background-color: #ffffff !important;&#10;}&#10;&#10;.active-part .nav-link-text {&#10; font-size: 0.9rem;&#10; font-weight: 700;&#10; color: #212529;&#10; text-decoration: underline;&#10; text-underline-offset: 3px;&#10;}&#10;&#10;.locked-part .nav-link-text {&#10; color: #ced4da;&#10;}&#10;&#10;.list-group-item {&#10; border-bottom: 1px solid #f8f9fa;&#10; padding: 1rem 1.25rem !important;&#10;}&#10;&lt;/style&gt;&#10;&#10;&lt;aside class="series-pull-quote shadow-sm"&gt;&#10; &lt;div class="series-header"&gt;&#10; READ MORE IN THIS SERIES&#10; &lt;/div&gt;&#10; &lt;div class="list-group list-group-flush"&gt;&#10; &#10; &#10; &#10;&#10; &lt;div class="list-group-item active-part "&gt;&#10; &lt;div class="d-flex justify-content-between align-items-center"&gt;&#10; &lt;div&gt;&#10; &#10; &lt;a href="https://www.sakurasky.com/blog/lakehouse-bronze-data-erasure-part-1/" class="nav-link-text"&gt;Part 1 - Keeping Raw Data You May Be Asked to Delete&lt;/a&gt;&#10; &#10; &lt;/div&gt;&#10;&#10; &lt;/div&gt;&#10; &lt;/div&gt;&#10; &#10; &#10; &#10;&#10; &lt;div class="list-group-item locked-part"&gt;&#10; &lt;div class="d-flex justify-content-between align-items-center"&gt;&#10; &lt;div&gt;&#10; &#10; &lt;span class="nav-link-text"&gt;Part 2 - Native Tables, Managed Iceberg, or Iceberg in an Open Catalog&lt;/span&gt;&#10; &#10; &lt;/div&gt;&#10;&#10; &lt;/div&gt;&#10; &lt;/div&gt;&#10; &#10; &#10; &#10;&#10; &lt;div class="list-group-item locked-part"&gt;&#10; &lt;div class="d-flex justify-content-between align-items-center"&gt;&#10; &lt;div&gt;&#10; &#10; &lt;span class="nav-link-text"&gt;Part 3 - Matching Records Without Rewriting History&lt;/span&gt;&#10; &#10; &lt;/div&gt;&#10;&#10; &lt;/div&gt;&#10; &lt;/div&gt;&#10; &#10; &#10; &#10;&#10; &lt;div class="list-group-item locked-part"&gt;&#10; &lt;div class="d-flex justify-content-between align-items-center"&gt;&#10; &lt;div&gt;&#10; &#10; &lt;span class="nav-link-text"&gt;Part 4 - Gold Tables With Terms of Use&lt;/span&gt;&#10; &#10; &lt;/div&gt;&#10;&#10; &lt;/div&gt;&#10; &lt;/div&gt;&#10; &#10; &#10; &#10;&#10; &lt;div class="list-group-item locked-part"&gt;&#10; &lt;div class="d-flex justify-content-between align-items-center"&gt;&#10; &lt;div&gt;&#10; &#10; &lt;span class="nav-link-text"&gt;Part 5 - When Yesterday&amp;#39;s Numbers Change&lt;/span&gt;&#10; &#10; &lt;/div&gt;&#10;&#10; &lt;/div&gt;&#10; &lt;/div&gt;&#10; &#10; &#10; &#10;&#10; &lt;div class="list-group-item locked-part"&gt;&#10; &lt;div class="d-flex justify-content-between align-items-center"&gt;&#10; &lt;div&gt;&#10; &#10; &lt;span class="nav-link-text"&gt;Part 6 - Which Failures Should Wake Someone Up&lt;/span&gt;&#10; &#10; &lt;/div&gt;&#10;&#10; &lt;/div&gt;&#10; &lt;/div&gt;&#10; &#10; &#10; &#10;&#10; &lt;div class="list-group-item locked-part"&gt;&#10; &lt;div class="d-flex justify-content-between align-items-center"&gt;&#10; &lt;div&gt;&#10; &#10; &lt;span class="nav-link-text"&gt;Part 7 - From an Entra Group to a BigQuery Row&lt;/span&gt;&#10; &#10; &lt;/div&gt;&#10;&#10; &lt;/div&gt;&#10; &lt;/div&gt;&#10; &#10; &#10; &#10;&#10; &lt;div class="list-group-item locked-part"&gt;&#10; &lt;div class="d-flex justify-content-between align-items-center"&gt;&#10; &lt;div&gt;&#10; &#10; &lt;span class="nav-link-text"&gt;Part 8 - Cost Levers, One at a Time&lt;/span&gt;&#10; &#10; &lt;/div&gt;&#10;&#10; &lt;/div&gt;&#10; &lt;/div&gt;&#10; &#10; &#10; &#10;&#10; &lt;div class="list-group-item locked-part"&gt;&#10; &lt;div class="d-flex justify-content-between align-items-center"&gt;&#10; &lt;div&gt;&#10; &#10; &lt;span class="nav-link-text"&gt;Part 9 - Rebuilding Gold After a Bad Day&lt;/span&gt;&#10; &#10; &lt;/div&gt;&#10;&#10; &lt;/div&gt;&#10; &lt;/div&gt;&#10; &#10; &lt;/div&gt;&#10;&lt;/aside&gt;&#10;&#10;&#10;&lt;p&gt;Your first job on an erasure request is a search: where does this person&amp;rsquo;s data sit? For a customer with one online order, the example estate in this series holds it in more than twenty places across the source database, the lakehouse, and the records around it. Some are obvious, such as the orders table and its bronze copy. Others are easy to miss: the change tables that feed replication, a month-end snapshot, the text of a query someone ran to look into a refund, and the table that caught a pipeline error. Each has its own rule for how long data stays after a delete, and several keep it for weeks after the DELETE statement succeeds.&lt;/p&gt;</description></item></channel></rss>