<?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 &amp;gt;&amp;gt;...I sent the project and in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950192#M15171</link>
    <description>&amp;gt;&amp;gt;...I sent the project and source code as an attachment, but I dont see it in the chat...

Why wouldn't you add a zip-file with the test project to a new post in the current thread?

In that case everybody will be able to test, if interested. As soon as you add the zip-file press 'Start upload' button and as soon as download completed press 'Submit' button.

Note: There is a little problem. A developer should have MKL v10 and MKL v11 installed on its computer. I don't have MKL v10, for example.</description>
    <pubDate>Mon, 25 Feb 2013 00:34:00 GMT</pubDate>
    <dc:creator>SergeyKostrov</dc:creator>
    <dc:date>2013-02-25T00:34:00Z</dc:date>
    <item>
      <title>MKL 11 Versus MKL 10 Performance</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950187#M15166</link>
      <description>&lt;P&gt;I have a benchmark application that mainly does two types of things:&lt;/P&gt;
&lt;P&gt;* FFTs&lt;/P&gt;
&lt;P&gt;* Eigenvector decompositions&lt;/P&gt;
&lt;P&gt;In both cases, the new MKL11 benchmarks much slower than MKL10 on a I7 IVY Bridge, Sandy Bridge and other CPUS.&lt;/P&gt;
&lt;P&gt;For the Eigenvector code, it is about 1/2 the performance in version 11.&amp;nbsp; For the FFTs the performance is degraded by about 30%.&lt;/P&gt;
&lt;P&gt;Is there a reason why MKL 11 performs worse?&amp;nbsp; Is there something I need to do to at least get equivalent performance to MKL 10?&lt;/P&gt;
&lt;P&gt;Thanks for your assistance in advance.&lt;/P&gt;</description>
      <pubDate>Wed, 20 Feb 2013 23:12:49 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950187#M15166</guid>
      <dc:creator>Jason_S_4</dc:creator>
      <dc:date>2013-02-20T23:12:49Z</dc:date>
    </item>
    <item>
      <title>Hello,</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950188#M15167</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;
&lt;P&gt;could you please a sample code show this performance problem?&amp;nbsp; Also let us know 1)how code is linked 2) Is it a 32 bit code or Intel64 code 3) Windows/Linux?&lt;/P&gt;
&lt;P&gt;Regards,&lt;BR /&gt;Chao&lt;/P&gt;</description>
      <pubDate>Thu, 21 Feb 2013 02:32:50 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950188#M15167</guid>
      <dc:creator>Chao_Y_Intel</dc:creator>
      <dc:date>2013-02-21T02:32:50Z</dc:date>
    </item>
    <item>
      <title>I have attached a complete</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950189#M15168</link>
      <description>&lt;P&gt;I have attached a complete VS2008 project which demonstrates the problem.&amp;nbsp; Press the "EigenValues" button and &amp;nbsp;"Run Benchmark" buttons to see the performance.&amp;nbsp; As a correction the FFT times are improved in version 11 BUT the eigenvalue calculation times are substantially worse under version 11.&amp;nbsp; Here is a summary of measured benchmarks:&lt;/P&gt;
&lt;P&gt;EIGENBenchmark:&lt;/P&gt;
&lt;P&gt;&amp;nbsp; I7-920 32Bit MKL10:&amp;nbsp;1.3 Seconds; I7-920 32Bit MKL11:&amp;nbsp;2.53 Seconds; I7-920 64Bit MKL10:&amp;nbsp;1.09 Seconds; I7-920 64Bit MKL11:&amp;nbsp;1.65 Seconds&lt;/P&gt;
&lt;P&gt;&amp;nbsp;I7-3770 32Bit MKL10: .757 Seconds; I7-3770 32Bit MKL11: 1.4 Seconds; I7-3770 64Bit MKL10: .63 Seconds, I7-3770 64Bit MKL1: .98 Seconds&lt;/P&gt;
&lt;P&gt;In order to compile, you copy the noted (in .txt file) .lib files into the folders for libraries by the way.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 21 Feb 2013 17:51:26 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950189#M15168</guid>
      <dc:creator>Jason_S_4</dc:creator>
      <dc:date>2013-02-21T17:51:26Z</dc:date>
    </item>
    <item>
      <title>So the mystery is: why is</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950190#M15169</link>
      <description>&lt;P&gt;So the mystery is: why is MKL11 substantially worse in performance for Eigenvector decompositions compared to MKL 10.&lt;/P&gt;
&lt;P&gt;Regards&lt;/P&gt;</description>
      <pubDate>Thu, 21 Feb 2013 17:53:08 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950190#M15169</guid>
      <dc:creator>Jason_S_4</dc:creator>
      <dc:date>2013-02-21T17:53:08Z</dc:date>
    </item>
    <item>
      <title>I sent the project and source</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950191#M15170</link>
      <description>&lt;P&gt;I sent the project and source code as an attachment, but I dont see it in the chat.&amp;nbsp; Is there some magic I need to do to get the attachment uploaded?&amp;nbsp; Should I email it somewhere?&lt;/P&gt;</description>
      <pubDate>Thu, 21 Feb 2013 17:56:13 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950191#M15170</guid>
      <dc:creator>Jason_S_4</dc:creator>
      <dc:date>2013-02-21T17:56:13Z</dc:date>
    </item>
    <item>
      <title>&gt;&gt;...I sent the project and</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950192#M15171</link>
      <description>&amp;gt;&amp;gt;...I sent the project and source code as an attachment, but I dont see it in the chat...

Why wouldn't you add a zip-file with the test project to a new post in the current thread?

In that case everybody will be able to test, if interested. As soon as you add the zip-file press 'Start upload' button and as soon as download completed press 'Submit' button.

Note: There is a little problem. A developer should have MKL v10 and MKL v11 installed on its computer. I don't have MKL v10, for example.</description>
      <pubDate>Mon, 25 Feb 2013 00:34:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-11-Versus-MKL-10-Performance/m-p/950192#M15171</guid>
      <dc:creator>SergeyKostrov</dc:creator>
      <dc:date>2013-02-25T00:34:00Z</dc:date>
    </item>
  </channel>
</rss>

