<?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 Re: Performance differences between IPP and MKL libraries in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/Performance-differences-between-IPP-and-MKL-libraries/m-p/910502#M14105</link>
    <description>&lt;P&gt;Hello,&lt;/P&gt;
&lt;P&gt;In general, the MKL library is intended to work with large data, where as IPP optimized to work with small amount of data. By the way, MKL uses IPP internally when data is not big, for example in FFT and DFT operations.&lt;/P&gt;
&lt;P&gt;You also might be interested to see the similar threads at this forum:&lt;/P&gt;
&lt;P&gt;&lt;A href="https://community.intel.com/ISN/Community/en-US/forums/permalink/5482247/5482233/ShowThread.aspx#5482233"&gt;Which FFT lib is faster, IPP or MKL?&lt;/A&gt;&lt;BR /&gt;&lt;A href="https://community.intel.com/ISN/Community/en-US/forums/permalink/5482468/5482448/ShowThread.aspx#5482448"&gt;IPP or MKL... or both?&lt;/A&gt;&lt;BR /&gt;&lt;A href="https://community.intel.com/ISN/Community/en-US/forums/permalink/5482480/5482480/ShowThread.aspx#5482480"&gt;Performance: IPP or MKL&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;Regards,&lt;BR /&gt; Vladimir&lt;/P&gt;</description>
    <pubDate>Wed, 20 Sep 2006 01:39:35 GMT</pubDate>
    <dc:creator>Vladimir_Dudnik</dc:creator>
    <dc:date>2006-09-20T01:39:35Z</dc:date>
    <item>
      <title>Performance differences between IPP and MKL libraries</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Performance-differences-between-IPP-and-MKL-libraries/m-p/910501#M14104</link>
      <description>&lt;P&gt;There is a substantial overlap in functionality between the IPP and the MKL libraries (for example, the axpy functionality, routines to compute the sin or cos of each element in a vector etc.) Does anyone have data to indicate whether there is a significant performance difference between these libraries?&lt;/P&gt;</description>
      <pubDate>Tue, 19 Sep 2006 06:33:41 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Performance-differences-between-IPP-and-MKL-libraries/m-p/910501#M14104</guid>
      <dc:creator>tbruegge</dc:creator>
      <dc:date>2006-09-19T06:33:41Z</dc:date>
    </item>
    <item>
      <title>Re: Performance differences between IPP and MKL libraries</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Performance-differences-between-IPP-and-MKL-libraries/m-p/910502#M14105</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;
&lt;P&gt;In general, the MKL library is intended to work with large data, where as IPP optimized to work with small amount of data. By the way, MKL uses IPP internally when data is not big, for example in FFT and DFT operations.&lt;/P&gt;
&lt;P&gt;You also might be interested to see the similar threads at this forum:&lt;/P&gt;
&lt;P&gt;&lt;A href="https://community.intel.com/ISN/Community/en-US/forums/permalink/5482247/5482233/ShowThread.aspx#5482233"&gt;Which FFT lib is faster, IPP or MKL?&lt;/A&gt;&lt;BR /&gt;&lt;A href="https://community.intel.com/ISN/Community/en-US/forums/permalink/5482468/5482448/ShowThread.aspx#5482448"&gt;IPP or MKL... or both?&lt;/A&gt;&lt;BR /&gt;&lt;A href="https://community.intel.com/ISN/Community/en-US/forums/permalink/5482480/5482480/ShowThread.aspx#5482480"&gt;Performance: IPP or MKL&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;Regards,&lt;BR /&gt; Vladimir&lt;/P&gt;</description>
      <pubDate>Wed, 20 Sep 2006 01:39:35 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Performance-differences-between-IPP-and-MKL-libraries/m-p/910502#M14105</guid>
      <dc:creator>Vladimir_Dudnik</dc:creator>
      <dc:date>2006-09-20T01:39:35Z</dc:date>
    </item>
  </channel>
</rss>

