<?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>question Re: Can I run the balancer for hdfs in Archives of Support Questions (Read Only)</title>
    <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Can-I-run-the-balancer-for-hdfs/m-p/78918#M82593</link>
    <description>There will not be any operational problems such as crashes or errors when&lt;BR /&gt;running a HDFS balancer on a cluster with HBase running, but there can&lt;BR /&gt;potentially be a performance impact depending on what the balancer decides&lt;BR /&gt;to move based on its space thresholds.&lt;BR /&gt;&lt;BR /&gt;The performance impact would come from loss of locality - the&lt;BR /&gt;RegionServers' required HFiles may find their blocks to be remote, so a&lt;BR /&gt;slightly higher network usage can be observed until the next major&lt;BR /&gt;compaction rewrites a block replica locally.&lt;BR /&gt;&lt;BR /&gt;If you'd like to narrow down the time-frame of impact, you can run the HDFS&lt;BR /&gt;balancer with the desired balancing threshold, and then once it is&lt;BR /&gt;complete, immediately follow up with a major compaction command on your&lt;BR /&gt;latency-sensitive HBase tables.</description>
    <pubDate>Fri, 24 Aug 2018 02:38:55 GMT</pubDate>
    <dc:creator>Harsh J</dc:creator>
    <dc:date>2018-08-24T02:38:55Z</dc:date>
    <item>
      <title>Can I run the balancer for hdfs?</title>
      <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Can-I-run-the-balancer-for-hdfs/m-p/78914#M82592</link>
      <description>&lt;P&gt;I use cloudera cdh 4.0.4.&lt;/P&gt;&lt;P&gt;I run balancing on Hbase.&lt;/P&gt;&lt;P&gt;However, I have 10 data nodes, and only 5 servers are being used as hbase region servers.&lt;/P&gt;&lt;P&gt;Data node imbalance has occurred.&lt;/P&gt;&lt;P&gt;Is there a possibility that Hbase will cause problems when balancing with Hadoop hdfs?&lt;/P&gt;</description>
      <pubDate>Fri, 16 Sep 2022 13:37:17 GMT</pubDate>
      <guid>https://community.cloudera.com/t5/Archives-of-Support-Questions/Can-I-run-the-balancer-for-hdfs/m-p/78914#M82592</guid>
      <dc:creator>avengers</dc:creator>
      <dc:date>2022-09-16T13:37:17Z</dc:date>
    </item>
    <item>
      <title>Re: Can I run the balancer for hdfs</title>
      <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Can-I-run-the-balancer-for-hdfs/m-p/78918#M82593</link>
      <description>There will not be any operational problems such as crashes or errors when&lt;BR /&gt;running a HDFS balancer on a cluster with HBase running, but there can&lt;BR /&gt;potentially be a performance impact depending on what the balancer decides&lt;BR /&gt;to move based on its space thresholds.&lt;BR /&gt;&lt;BR /&gt;The performance impact would come from loss of locality - the&lt;BR /&gt;RegionServers' required HFiles may find their blocks to be remote, so a&lt;BR /&gt;slightly higher network usage can be observed until the next major&lt;BR /&gt;compaction rewrites a block replica locally.&lt;BR /&gt;&lt;BR /&gt;If you'd like to narrow down the time-frame of impact, you can run the HDFS&lt;BR /&gt;balancer with the desired balancing threshold, and then once it is&lt;BR /&gt;complete, immediately follow up with a major compaction command on your&lt;BR /&gt;latency-sensitive HBase tables.</description>
      <pubDate>Fri, 24 Aug 2018 02:38:55 GMT</pubDate>
      <guid>https://community.cloudera.com/t5/Archives-of-Support-Questions/Can-I-run-the-balancer-for-hdfs/m-p/78918#M82593</guid>
      <dc:creator>Harsh J</dc:creator>
      <dc:date>2018-08-24T02:38:55Z</dc:date>
    </item>
  </channel>
</rss>

