<?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 Too low memory bandwidth. in Software Tuning, Performance Optimization &amp; Platform Monitoring</title>
    <link>https://community.intel.com/t5/Software-Tuning-Performance/Too-low-memory-bandwidth/m-p/1134094#M6455</link>
    <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;I have made the program that conduct 10 independent pointer chase and I have verified 99% of pointer chase step goes to main memory.&lt;/P&gt;&lt;P&gt;My system has maximum memory bandwidth as&amp;nbsp;59.61GB/s&lt;/P&gt;&lt;P&gt;When I use memory bandwidth monitoring program, one thread can generate about 10GB/s memory bandwidth.&lt;/P&gt;&lt;P&gt;When I increase the number of process, memory bandwidth that each process generate is reduced and I have added&amp;nbsp;all the memory bandwidth value that each process generate, I&amp;nbsp;can only obtain&amp;nbsp;about 32GB/s memory bandwidth.&lt;/P&gt;&lt;P&gt;I can't get more than 32GB/s memory bandwidth even though I increase the number of thread.&lt;/P&gt;&lt;P&gt;So, I think this result comes from bank conflict so leads to&amp;nbsp;low utilization of memory bandwidth.&lt;/P&gt;&lt;P&gt;Is this reasonable explanation?? or Should I consider another factor??&lt;/P&gt;&lt;P&gt;Thank you.&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Thu, 25 Jul 2019 09:59:14 GMT</pubDate>
    <dc:creator>Jaeyoung__Choi</dc:creator>
    <dc:date>2019-07-25T09:59:14Z</dc:date>
    <item>
      <title>Too low memory bandwidth.</title>
      <link>https://community.intel.com/t5/Software-Tuning-Performance/Too-low-memory-bandwidth/m-p/1134094#M6455</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;I have made the program that conduct 10 independent pointer chase and I have verified 99% of pointer chase step goes to main memory.&lt;/P&gt;&lt;P&gt;My system has maximum memory bandwidth as&amp;nbsp;59.61GB/s&lt;/P&gt;&lt;P&gt;When I use memory bandwidth monitoring program, one thread can generate about 10GB/s memory bandwidth.&lt;/P&gt;&lt;P&gt;When I increase the number of process, memory bandwidth that each process generate is reduced and I have added&amp;nbsp;all the memory bandwidth value that each process generate, I&amp;nbsp;can only obtain&amp;nbsp;about 32GB/s memory bandwidth.&lt;/P&gt;&lt;P&gt;I can't get more than 32GB/s memory bandwidth even though I increase the number of thread.&lt;/P&gt;&lt;P&gt;So, I think this result comes from bank conflict so leads to&amp;nbsp;low utilization of memory bandwidth.&lt;/P&gt;&lt;P&gt;Is this reasonable explanation?? or Should I consider another factor??&lt;/P&gt;&lt;P&gt;Thank you.&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 25 Jul 2019 09:59:14 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Tuning-Performance/Too-low-memory-bandwidth/m-p/1134094#M6455</guid>
      <dc:creator>Jaeyoung__Choi</dc:creator>
      <dc:date>2019-07-25T09:59:14Z</dc:date>
    </item>
  </channel>
</rss>

