<?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 VTune uses hardware PMU to in Analyzers</title>
    <link>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089593#M15564</link>
    <description>&lt;P&gt;VTune uses hardware PMU to measure. The general formula is -&lt;/P&gt;

&lt;P&gt;MFLOPS Formula = A-specific-event counters / 1,000,000 / Elapsed Time&lt;BR /&gt;
	Elapsed time = CPU_CLK_UNHALTED.THREAD / Processor-Frequency / Number-of-Cores&lt;/P&gt;

&lt;P&gt;Please see this&lt;A href="https://software.intel.com/en-us/blogs/2015/04/20/use-which-hardware-pmu-events-to-calculate-flops-on-intelr-xeon-phitm-coprocessor?wapkw=VTune%20gflops%20PMU%20more%3Ap%3AtechArticle-keywords%3A*%20OR%20more%3Ap%3Ablog-keywords%3A*%20OR%20more%3Ap%3AvideoObject-keywords%3A*%20OR%20more%3Ap%3AdataCatalog-keywords%3A*%20OR%20more%3Ap%3Abook-keywords%3A*%20OR%20more%3Ap%3AsoftwareApplication-keywords%3A*%20OR%20more%3Ap%3Acode-keywords%3A*%20OR%20more%3Ap%3Aquestion-keywords%3A*"&gt; blog&lt;/A&gt; to know examples in detail.&lt;/P&gt;</description>
    <pubDate>Sat, 16 Jan 2016 10:47:38 GMT</pubDate>
    <dc:creator>Peter_W_Intel</dc:creator>
    <dc:date>2016-01-16T10:47:38Z</dc:date>
    <item>
      <title>How does VTune collect HPC data?</title>
      <link>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089592#M15563</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;I am new to VTune and was interested in knowing how does VTune collect the performance counter data?&amp;nbsp;&lt;SPAN class="s1"&gt;Does vTune capture counters while the OS code is executing? Are the counters being updated in user space or in kernel space?&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;Thanks.&lt;/P&gt;</description>
      <pubDate>Sat, 16 Jan 2016 09:23:14 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089592#M15563</guid>
      <dc:creator>Baljit_S_</dc:creator>
      <dc:date>2016-01-16T09:23:14Z</dc:date>
    </item>
    <item>
      <title>VTune uses hardware PMU to</title>
      <link>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089593#M15564</link>
      <description>&lt;P&gt;VTune uses hardware PMU to measure. The general formula is -&lt;/P&gt;

&lt;P&gt;MFLOPS Formula = A-specific-event counters / 1,000,000 / Elapsed Time&lt;BR /&gt;
	Elapsed time = CPU_CLK_UNHALTED.THREAD / Processor-Frequency / Number-of-Cores&lt;/P&gt;

&lt;P&gt;Please see this&lt;A href="https://software.intel.com/en-us/blogs/2015/04/20/use-which-hardware-pmu-events-to-calculate-flops-on-intelr-xeon-phitm-coprocessor?wapkw=VTune%20gflops%20PMU%20more%3Ap%3AtechArticle-keywords%3A*%20OR%20more%3Ap%3Ablog-keywords%3A*%20OR%20more%3Ap%3AvideoObject-keywords%3A*%20OR%20more%3Ap%3AdataCatalog-keywords%3A*%20OR%20more%3Ap%3Abook-keywords%3A*%20OR%20more%3Ap%3AsoftwareApplication-keywords%3A*%20OR%20more%3Ap%3Acode-keywords%3A*%20OR%20more%3Ap%3Aquestion-keywords%3A*"&gt; blog&lt;/A&gt; to know examples in detail.&lt;/P&gt;</description>
      <pubDate>Sat, 16 Jan 2016 10:47:38 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089593#M15564</guid>
      <dc:creator>Peter_W_Intel</dc:creator>
      <dc:date>2016-01-16T10:47:38Z</dc:date>
    </item>
    <item>
      <title>Hello,</title>
      <link>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089594#M15565</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;It is worth also to look at &lt;A href="https://software.intel.com/en-us/node/544067"&gt;https://software.intel.com/en-us/node/544067&lt;/A&gt; to understand the sampling collection method that VTune use for hardware (driver) based collections.&lt;/P&gt;

&lt;P&gt;BTW - what particular HPC data are you interested in?&lt;/P&gt;

&lt;P&gt;Thanks &amp;amp; Regards, Dmitry&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 18 Jan 2016 08:13:51 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089594#M15565</guid>
      <dc:creator>Dmitry_P_Intel1</dc:creator>
      <dc:date>2016-01-18T08:13:51Z</dc:date>
    </item>
    <item>
      <title>And, actually, there are two</title>
      <link>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089595#M15566</link>
      <description>&lt;P&gt;And, actually, there are two different collection mechanisms: one is software-based and one is hardware-based. &amp;nbsp;It depends on the analysis type you select.&lt;/P&gt;

&lt;P&gt;So, again, what specifically are you asking about?&lt;/P&gt;</description>
      <pubDate>Tue, 19 Jan 2016 18:22:10 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089595#M15566</guid>
      <dc:creator>David_A_Intel1</dc:creator>
      <dc:date>2016-01-19T18:22:10Z</dc:date>
    </item>
    <item>
      <title>Thanks for the reply guys. I</title>
      <link>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089596#M15567</link>
      <description>&lt;P&gt;Thanks for the reply guys.&amp;nbsp;&lt;SPAN style="font-size: 1em;"&gt;I am dealing with HPC for branches, cache misses and memory uops.&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Tue, 19 Jan 2016 20:13:31 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/How-does-VTune-collect-HPC-data/m-p/1089596#M15567</guid>
      <dc:creator>Baljit_S_</dc:creator>
      <dc:date>2016-01-19T20:13:31Z</dc:date>
    </item>
  </channel>
</rss>

