<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Why doesn't my job-cluster scale down? in missing-QuestionPost</title>
    <link>https://community.databricks.com/t5/missing-questionpost/why-doesn-t-my-job-cluster-scale-down/m-p/2773#M17</link>
    <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;Running on Databricks-AWS, I have a job running on a cluster with 3 workers, 2-cores each (r6i.large), with &lt;B&gt;autoscaling enabled&lt;/B&gt;.&lt;/P&gt;&lt;P&gt;&lt;U&gt;The Spark job has two stages&lt;/U&gt;: &lt;/P&gt;&lt;P&gt;&lt;B&gt;(1)&lt;/B&gt; highly parallelizable, cpu-intensive stage. This stage takes 15 minutes.&lt;/P&gt;&lt;P&gt;&lt;B&gt;(2)&lt;/B&gt; a non-parallelizable stage (only a single partition, so a single spark task). This stage takes 45 minutes.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;In the first stage, the cluster scales up from 1 worker to 3, and all 3 workers (6 cores) are fully utilized for the duration of the stage (15 minutes). Then, in the second stage, only a single worker node is active for the entire 45 minutes, but databricks does not scale down my cluster and I have two nodes completely idle for 45 minutes.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Any idea why that is, and how I can utilize autoscaling to be more cost efficient in this type of job?&lt;/P&gt;&lt;P&gt;Thanks!&lt;/P&gt;</description>
    <pubDate>Wed, 21 Jun 2023 12:48:27 GMT</pubDate>
    <dc:creator>804925</dc:creator>
    <dc:date>2023-06-21T12:48:27Z</dc:date>
    <item>
      <title>Why doesn't my job-cluster scale down?</title>
      <link>https://community.databricks.com/t5/missing-questionpost/why-doesn-t-my-job-cluster-scale-down/m-p/2773#M17</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;Running on Databricks-AWS, I have a job running on a cluster with 3 workers, 2-cores each (r6i.large), with &lt;B&gt;autoscaling enabled&lt;/B&gt;.&lt;/P&gt;&lt;P&gt;&lt;U&gt;The Spark job has two stages&lt;/U&gt;: &lt;/P&gt;&lt;P&gt;&lt;B&gt;(1)&lt;/B&gt; highly parallelizable, cpu-intensive stage. This stage takes 15 minutes.&lt;/P&gt;&lt;P&gt;&lt;B&gt;(2)&lt;/B&gt; a non-parallelizable stage (only a single partition, so a single spark task). This stage takes 45 minutes.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;In the first stage, the cluster scales up from 1 worker to 3, and all 3 workers (6 cores) are fully utilized for the duration of the stage (15 minutes). Then, in the second stage, only a single worker node is active for the entire 45 minutes, but databricks does not scale down my cluster and I have two nodes completely idle for 45 minutes.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Any idea why that is, and how I can utilize autoscaling to be more cost efficient in this type of job?&lt;/P&gt;&lt;P&gt;Thanks!&lt;/P&gt;</description>
      <pubDate>Wed, 21 Jun 2023 12:48:27 GMT</pubDate>
      <guid>https://community.databricks.com/t5/missing-questionpost/why-doesn-t-my-job-cluster-scale-down/m-p/2773#M17</guid>
      <dc:creator>804925</dc:creator>
      <dc:date>2023-06-21T12:48:27Z</dc:date>
    </item>
    <item>
      <title>Re: Why doesn't my job-cluster scale down?</title>
      <link>https://community.databricks.com/t5/missing-questionpost/why-doesn-t-my-job-cluster-scale-down/m-p/2774#M18</link>
      <description>&lt;P&gt;Hi @Yoav Ben​&amp;nbsp;&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Great to meet you, and thanks for your question! &lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Let's see if your peers in the community have an answer to your question. Thanks.&lt;/P&gt;</description>
      <pubDate>Thu, 22 Jun 2023 05:03:42 GMT</pubDate>
      <guid>https://community.databricks.com/t5/missing-questionpost/why-doesn-t-my-job-cluster-scale-down/m-p/2774#M18</guid>
      <dc:creator>Anonymous</dc:creator>
      <dc:date>2023-06-22T05:03:42Z</dc:date>
    </item>
  </channel>
</rss>

