<?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 Endpoint performance questions in Machine Learning</title>
    <link>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63303#M3094</link>
    <description>&lt;P&gt;Hi!&amp;nbsp;&lt;BR /&gt;Had really interesting results from some endpoint performance tests I did. I set up the non-optimized endpoint with zero-cluster scaling and optimized had this feature disabled.&lt;BR /&gt;&lt;BR /&gt;1) Why does the non-optimized endpoint have variable response time for 3600, 1800, and 600 seconds tests? If the serving cluster node scaled to 0 (due to no traffic) I would expect it to also require 240 seconds to start up and start serving again.&amp;nbsp;&lt;/P&gt;&lt;P&gt;- what is going on behind the scenes that results in this?&lt;BR /&gt;&lt;BR /&gt;2) It was also interesting to see that the endpoint metrcs showed request error rates (top right graph). The endpoint didnt have any bad responses. Also the logs didnt show anything that would allude to this. Any idea why this would be the case? See blow for the metrics image.&lt;/P&gt;&lt;P&gt;3) I didnt find much information on this on the databricks documentation. Any additional documentation would be appreicated! Happy to sync with the team&lt;BR /&gt;&lt;BR /&gt;non-optimized endpoint results&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Kaizen_1-1710196442817.png" style="width: 400px;"&gt;&lt;img src="https://community.databricks.com/t5/image/serverpage/image-id/6610i2F8785428A71BEA3/image-size/medium/is-moderation-mode/true?v=v2&amp;amp;px=400" role="button" title="Kaizen_1-1710196442817.png" alt="Kaizen_1-1710196442817.png" /&gt;&lt;/span&gt;&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;optimized endpoint results&amp;nbsp;&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Kaizen_0-1710196408535.png" style="width: 400px;"&gt;&lt;img src="https://community.databricks.com/t5/image/serverpage/image-id/6609i7F64DB19537F3D1B/image-size/medium/is-moderation-mode/true?v=v2&amp;amp;px=400" role="button" title="Kaizen_0-1710196408535.png" alt="Kaizen_0-1710196408535.png" /&gt;&lt;/span&gt;&lt;/P&gt;&lt;P&gt;metrics log:&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Kaizen_2-1710196880601.png" style="width: 400px;"&gt;&lt;img src="https://community.databricks.com/t5/image/serverpage/image-id/6611iF5130CE717D5A1A3/image-size/medium/is-moderation-mode/true?v=v2&amp;amp;px=400" role="button" title="Kaizen_2-1710196880601.png" alt="Kaizen_2-1710196880601.png" /&gt;&lt;/span&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Mon, 11 Mar 2024 22:47:05 GMT</pubDate>
    <dc:creator>Kaizen</dc:creator>
    <dc:date>2024-03-11T22:47:05Z</dc:date>
    <item>
      <title>Endpoint performance questions</title>
      <link>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63303#M3094</link>
      <description>&lt;P&gt;Hi!&amp;nbsp;&lt;BR /&gt;Had really interesting results from some endpoint performance tests I did. I set up the non-optimized endpoint with zero-cluster scaling and optimized had this feature disabled.&lt;BR /&gt;&lt;BR /&gt;1) Why does the non-optimized endpoint have variable response time for 3600, 1800, and 600 seconds tests? If the serving cluster node scaled to 0 (due to no traffic) I would expect it to also require 240 seconds to start up and start serving again.&amp;nbsp;&lt;/P&gt;&lt;P&gt;- what is going on behind the scenes that results in this?&lt;BR /&gt;&lt;BR /&gt;2) It was also interesting to see that the endpoint metrcs showed request error rates (top right graph). The endpoint didnt have any bad responses. Also the logs didnt show anything that would allude to this. Any idea why this would be the case? See blow for the metrics image.&lt;/P&gt;&lt;P&gt;3) I didnt find much information on this on the databricks documentation. Any additional documentation would be appreicated! Happy to sync with the team&lt;BR /&gt;&lt;BR /&gt;non-optimized endpoint results&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Kaizen_1-1710196442817.png" style="width: 400px;"&gt;&lt;img src="https://community.databricks.com/t5/image/serverpage/image-id/6610i2F8785428A71BEA3/image-size/medium/is-moderation-mode/true?v=v2&amp;amp;px=400" role="button" title="Kaizen_1-1710196442817.png" alt="Kaizen_1-1710196442817.png" /&gt;&lt;/span&gt;&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;optimized endpoint results&amp;nbsp;&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Kaizen_0-1710196408535.png" style="width: 400px;"&gt;&lt;img src="https://community.databricks.com/t5/image/serverpage/image-id/6609i7F64DB19537F3D1B/image-size/medium/is-moderation-mode/true?v=v2&amp;amp;px=400" role="button" title="Kaizen_0-1710196408535.png" alt="Kaizen_0-1710196408535.png" /&gt;&lt;/span&gt;&lt;/P&gt;&lt;P&gt;metrics log:&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Kaizen_2-1710196880601.png" style="width: 400px;"&gt;&lt;img src="https://community.databricks.com/t5/image/serverpage/image-id/6611iF5130CE717D5A1A3/image-size/medium/is-moderation-mode/true?v=v2&amp;amp;px=400" role="button" title="Kaizen_2-1710196880601.png" alt="Kaizen_2-1710196880601.png" /&gt;&lt;/span&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 11 Mar 2024 22:47:05 GMT</pubDate>
      <guid>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63303#M3094</guid>
      <dc:creator>Kaizen</dc:creator>
      <dc:date>2024-03-11T22:47:05Z</dc:date>
    </item>
    <item>
      <title>Re: Endpoint performance questions</title>
      <link>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63305#M3095</link>
      <description>&lt;P&gt;&lt;a href="https://community.databricks.com/t5/user/viewprofilepage/user-id/29"&gt;@s_park&lt;/a&gt;&amp;nbsp;/&amp;nbsp;&lt;a href="https://community.databricks.com/t5/user/viewprofilepage/user-id/5"&gt;@Sujitha&lt;/a&gt;&amp;nbsp;/&amp;nbsp;&lt;a href="https://community.databricks.com/t5/user/viewprofilepage/user-id/26078"&gt;@Debayan&lt;/a&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 11 Mar 2024 22:45:36 GMT</pubDate>
      <guid>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63305#M3095</guid>
      <dc:creator>Kaizen</dc:creator>
      <dc:date>2024-03-11T22:45:36Z</dc:date>
    </item>
    <item>
      <title>Re: Endpoint performance questions</title>
      <link>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63306#M3096</link>
      <description>&lt;P&gt;Answering Q1:&amp;nbsp;&lt;BR /&gt;1) The variable response time is due to the first endpoint response time requiring ~180 seconds to scale to 1 cluster from 0&lt;BR /&gt;&lt;BR /&gt;2) Can i change zero scale time from the preset 30 min?&lt;/P&gt;</description>
      <pubDate>Mon, 11 Mar 2024 22:55:49 GMT</pubDate>
      <guid>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63306#M3096</guid>
      <dc:creator>Kaizen</dc:creator>
      <dc:date>2024-03-11T22:55:49Z</dc:date>
    </item>
    <item>
      <title>Re: Endpoint performance questions</title>
      <link>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63403#M3101</link>
      <description>&lt;P&gt;Thanks for this.&amp;nbsp;&lt;/P&gt;&lt;P&gt;1) The odd values i got for 3600/1800/ etc was due to an outlier in my data so in general a response time of ~183 sec should be expected&amp;nbsp;&lt;BR /&gt;&lt;BR /&gt;2)&amp;nbsp;&lt;a href="https://community.databricks.com/t5/user/viewprofilepage/user-id/9"&gt;@Retired_mod&lt;/a&gt;&amp;nbsp;can we adjust the scaling of the cluster from 30 min to something else?&lt;/P&gt;</description>
      <pubDate>Tue, 12 Mar 2024 15:19:44 GMT</pubDate>
      <guid>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63403#M3101</guid>
      <dc:creator>Kaizen</dc:creator>
      <dc:date>2024-03-12T15:19:44Z</dc:date>
    </item>
    <item>
      <title>Re: Endpoint performance questions</title>
      <link>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63720#M3114</link>
      <description>&lt;P&gt;&lt;A href="https://community.databricks.com/t5/user/viewprofilepage/user-id/29" target="_blank"&gt;@s_park&lt;/A&gt;&lt;SPAN&gt;&amp;nbsp;/&amp;nbsp;&lt;/SPAN&gt;&lt;A href="https://community.databricks.com/t5/user/viewprofilepage/user-id/5" target="_blank"&gt;@Sujitha&lt;/A&gt;&lt;SPAN&gt;&amp;nbsp;/&amp;nbsp;&lt;/SPAN&gt;&lt;A href="https://community.databricks.com/t5/user/viewprofilepage/user-id/26078" target="_blank"&gt;@Debayan&lt;/A&gt;&lt;SPAN&gt;&amp;nbsp; could one of you address item 2?&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Thu, 14 Mar 2024 17:16:06 GMT</pubDate>
      <guid>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/63720#M3114</guid>
      <dc:creator>Kaizen</dc:creator>
      <dc:date>2024-03-14T17:16:06Z</dc:date>
    </item>
    <item>
      <title>Re: Endpoint performance questions</title>
      <link>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/65921#M3181</link>
      <description>&lt;P&gt;Independently found the solution to item 2. Currently you cannot modify the 30 min time for scale to zero.&amp;nbsp;&lt;BR /&gt;&lt;BR /&gt;Hope this helps someone in the future!&lt;/P&gt;</description>
      <pubDate>Tue, 09 Apr 2024 16:52:57 GMT</pubDate>
      <guid>https://community.databricks.com/t5/machine-learning/endpoint-performance-questions/m-p/65921#M3181</guid>
      <dc:creator>Kaizen</dc:creator>
      <dc:date>2024-04-09T16:52:57Z</dc:date>
    </item>
  </channel>
</rss>

