<?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 amplifier options in Analyzers</title>
    <link>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811222#M1895</link>
    <description>Here is Peter's suggestion expanded more explicitly (without the -start-paused)&lt;BR /&gt;&lt;BR /&gt;In one terminal, run&lt;BR /&gt;&amp;gt; amplxe-cl -collect nehalem_general-exploration -duration 300&lt;BR /&gt;(or set the duration for however many seconds the run will take)&lt;BR /&gt;&lt;BR /&gt;In another terminal, perform the MPI run&lt;BR /&gt;&amp;gt; mpiexec ....&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;If you don't know how long the run will take, there are some alternate methods for stopping collection&lt;BR /&gt;&lt;BR /&gt;In one terminal, run&lt;BR /&gt;&amp;gt; amplxe-cl -collect nehalem_general-exploration -duration unlimited&lt;BR /&gt;&lt;BR /&gt;In another terminal, perform the MPI run&lt;BR /&gt;&amp;gt; mpiexec ...&lt;BR /&gt;&lt;BR /&gt;To stop collection, enter Ctrl-C in the first terminal, or use the 'stop' command:&lt;BR /&gt;&amp;gt; amplxe-cl -command stop&lt;BR /&gt;</description>
    <pubDate>Mon, 14 Feb 2011 16:57:10 GMT</pubDate>
    <dc:creator>Mark_D_Intel</dc:creator>
    <dc:date>2011-02-14T16:57:10Z</dc:date>
    <item>
      <title>vtune amplifier options</title>
      <link>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811220#M1893</link>
      <description>Hi,&lt;BR /&gt;we use vtune amplifier with IntelMPI 4.0 on nehalem architecture.&lt;BR /&gt;I load the module sep and pax like this : &lt;BR /&gt;Module Size Used by&lt;BR /&gt;sep3_1 56496 0 &lt;BR /&gt;pax 22368 0 &lt;BR /&gt;&lt;BR /&gt;My user is in vtune group.&lt;BR /&gt;The hotspots collect works but the options like nehalem_general-exploration are problematic.&lt;BR /&gt;&lt;BR /&gt;The process on compute node are :&lt;BR /&gt;gassowsk 15639 5329 0 15:49 ? 00:00:00 /opt/lsf/7.0/linux2.6-glibc2.3-x86_64/etc/res -d /opt/lsf/conf -m service2 /tmp_user/stelvio/lsf_spool/1296485356.302046&lt;BR /&gt;gassowsk 15641 15639 0 15:49 ? 00:00:00 /bin/sh /tmp_user/stelvio/lsf_spool/1296485356.302046&lt;BR /&gt;gassowsk 15645 15641 0 15:49 ? 00:00:00 /bin/sh /tmp_user/stelvio/lsf_spool/1296485356.302046.shell&lt;BR /&gt;gassowsk 15654 1 0 15:49 ? 00:00:00 python /tmp_user/stelvio/opt/intel/impi/3.2.1.009/bin/mpd&lt;BR /&gt;gassowsk 15662 15645 0 15:49 ? 00:00:00 amplxe-cl -collect nehalem_general-exploration -- mpiexec.hydra -bootstrap fork -np 8 /tmp_user/stelvio/gassowsk/TEST/gather.x&lt;BR /&gt;gassowsk 15671 15662 0 15:49 ? 00:00:00 /opt/intel/vtune_amplifier_xe_2011/bin64/amplxe-runsa --ui-output-format xml --ui-output-fd 4 --option-file /tmp_user/stelvio/gassowsk/TEST/r032ge/config/runsa.options&lt;BR /&gt;gassowsk 15707 15671 0 15:49 ? 00:00:00 /opt/intel/vtune_amplifier_xe_2011/bin64/amplxe-runsa --ui-output-format xml --ui-output-fd 4 --option-file /tmp_user/stelvio/gassowsk/TEST/r032ge/config/runsa.options&lt;BR /&gt;&lt;BR /&gt;The job don't terminate and is killed with cputime limit.&lt;BR /&gt;Thank you for your help.&lt;BR /&gt;</description>
      <pubDate>Tue, 01 Feb 2011 09:45:31 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811220#M1893</guid>
      <dc:creator>Pierre_B_1</dc:creator>
      <dc:date>2011-02-01T09:45:31Z</dc:date>
    </item>
    <item>
      <title>vtune amplifier options</title>
      <link>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811221#M1894</link>
      <description>VTune Amplifier XE 2011 can only collect performance data on one node. You have to install the product on other nodes separately,and collect performance data from different node.&lt;BR /&gt;&lt;BR /&gt;SoI suggest you to start data collection by using amplxe-cl but set start-paused, set duration time for all nodes. SecondaryrunMPI application without VTune Amplifier XE from one node, you can insert __itt code to do resume/paused control in MPI application (seethis &lt;A href="http://software.intel.com/en-us/articles/use-new-pause-and-resume-api-from-intel-vtune-amplifier-2011-xe/"&gt;article&lt;/A&gt;).Again, you have to set duration to terminate data collectionon all nodes.&lt;BR /&gt;&lt;BR /&gt;If you want tocollect performance data of MPI application on singlenode with multicore, please refer to this &lt;A href="http://software.intel.com/en-us/articles/why-cant-hotspots-analysis-trace-spawned-processes-of-mpi-job/"&gt;article&lt;/A&gt;.&lt;BR /&gt;&lt;BR /&gt;Regards, Peter</description>
      <pubDate>Thu, 10 Feb 2011 06:03:59 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811221#M1894</guid>
      <dc:creator>Peter_W_Intel</dc:creator>
      <dc:date>2011-02-10T06:03:59Z</dc:date>
    </item>
    <item>
      <title>vtune amplifier options</title>
      <link>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811222#M1895</link>
      <description>Here is Peter's suggestion expanded more explicitly (without the -start-paused)&lt;BR /&gt;&lt;BR /&gt;In one terminal, run&lt;BR /&gt;&amp;gt; amplxe-cl -collect nehalem_general-exploration -duration 300&lt;BR /&gt;(or set the duration for however many seconds the run will take)&lt;BR /&gt;&lt;BR /&gt;In another terminal, perform the MPI run&lt;BR /&gt;&amp;gt; mpiexec ....&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;If you don't know how long the run will take, there are some alternate methods for stopping collection&lt;BR /&gt;&lt;BR /&gt;In one terminal, run&lt;BR /&gt;&amp;gt; amplxe-cl -collect nehalem_general-exploration -duration unlimited&lt;BR /&gt;&lt;BR /&gt;In another terminal, perform the MPI run&lt;BR /&gt;&amp;gt; mpiexec ...&lt;BR /&gt;&lt;BR /&gt;To stop collection, enter Ctrl-C in the first terminal, or use the 'stop' command:&lt;BR /&gt;&amp;gt; amplxe-cl -command stop&lt;BR /&gt;</description>
      <pubDate>Mon, 14 Feb 2011 16:57:10 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811222#M1895</guid>
      <dc:creator>Mark_D_Intel</dc:creator>
      <dc:date>2011-02-14T16:57:10Z</dc:date>
    </item>
    <item>
      <title>vtune amplifier options</title>
      <link>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811223#M1896</link>
      <description>Ok thanks for those suggestions.&lt;BR /&gt;I made a lot of tests, I re-installed the soft et the driver.&lt;BR /&gt;I think there was conflicts or side effects between the first installation and the method to launch amplifier&lt;BR /&gt;on diskless nodes (cluster SGI ICE) with login ldap, batch server LSF, MPI libraries (Intel or MPT), etc ...... &lt;BR /&gt;&lt;BR /&gt;On one node everythings works !&lt;BR /&gt;Now my problem is to find a methodology to launch amplifier-cl on multinodes and to process the results in&lt;BR /&gt;a uniq amplifier gui. &lt;BR /&gt;&lt;BR /&gt;We would like to estimate the Flops, is it possible ?&lt;BR /&gt;&lt;BR /&gt;Best regards</description>
      <pubDate>Tue, 22 Feb 2011 14:57:34 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811223#M1896</guid>
      <dc:creator>Pierre_B_1</dc:creator>
      <dc:date>2011-02-22T14:57:34Z</dc:date>
    </item>
    <item>
      <title>vtune amplifier options</title>
      <link>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811224#M1897</link>
      <description>Sounds good - it works after uninstalling/reinstalling the product.&lt;BR /&gt;&lt;BR /&gt;There is no predefined analysis type for Flops, you can define (create) yourself new analysis typeto measure Flops: &lt;BR /&gt;1) On left-bottom of "Choose Analysis Type" dialog, click "New..." button to select "Hardware Event-based Sampling Analysis"&lt;BR /&gt;2) Edit new analysis type name&lt;BR /&gt;3) Click "Edit" button then click "Add Event" button to add new event(s) to your new analysis type. You might select event "X87_OPS_RETIRED.ANY" or/and event "FP_ASSIST"&lt;BR /&gt;&lt;BR /&gt;Regards, Peter</description>
      <pubDate>Wed, 23 Feb 2011 02:47:47 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811224#M1897</guid>
      <dc:creator>Peter_W_Intel</dc:creator>
      <dc:date>2011-02-23T02:47:47Z</dc:date>
    </item>
    <item>
      <title>vtune amplifier options</title>
      <link>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811225#M1898</link>
      <description>Hi Paul,&lt;BR /&gt;I just posted an article on how to estimate FLOPS using event based sampling. You might want to take a look at it.&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://origin-software.intel.com/en-us/articles/estimating-flops-using-event-based-sampling-ebs/" target="_blank"&gt;http://origin-software.intel.com/en-us/articles/estimating-flops-using-event-based-sampling-ebs/&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;Thanks&lt;BR /&gt;Levent</description>
      <pubDate>Thu, 03 Mar 2011 16:06:56 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/vtune-amplifier-options/m-p/811225#M1898</guid>
      <dc:creator>Huseyin_A_Intel1</dc:creator>
      <dc:date>2011-03-03T16:06:56Z</dc:date>
    </item>
  </channel>
</rss>

