<?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 Basic Code of Using MKL FFT on MIC in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Basic-Code-of-Using-MKL-FFT-on-MIC/m-p/1019980#M19675</link>
    <description>&lt;PRE class="brush:cpp;"&gt;&amp;nbsp;&lt;/PRE&gt;

&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;In our local cluster we have a bunch of MIC's, which are (almost) never esed. I would like to give it the try, but I have no experience iwth the MKL library or MIC. My programs are very simple and they are based on FFT:&lt;/P&gt;

&lt;P&gt;I have an iterative process in which the smae update procedure is applied to the previous data:&lt;/P&gt;

&lt;PRE class="brush:cpp;"&gt;// initialize data rho[lx]

for(t=0;t&amp;lt;tmax;t++)
{

// first step in real space is local
for(i=0;i&amp;lt;lx;i++)
{
nt(i)=rho&lt;I&gt;*rho&lt;I&gt;*rho&lt;I&gt;;
}
//forward FFT rho-&amp;gt;rhok
//forward FFT nt-&amp;gt; ntk

//update in k space 
for(i=0;i&amp;lt;lx/2+1;i++)
{
newrhok(i)=filter1(i)*rhok(i)+filter2(i)*ntk(i); // complex multiplication
}
//inverse FFT newrhok-&amp;gt;newrho
&lt;/I&gt;&lt;/I&gt;&lt;/I&gt;&lt;/PRE&gt;

&lt;P&gt;So my problem is how to ues MKL FFT's on MIC's and how minimize the transfers between cpu and MIC.&lt;/P&gt;

&lt;P&gt;I hope this is the right forum and many thanks in advance&lt;/P&gt;

&lt;P&gt;Cristian&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Thu, 11 Dec 2014 12:06:42 GMT</pubDate>
    <dc:creator>Cristian_Vasile_A_</dc:creator>
    <dc:date>2014-12-11T12:06:42Z</dc:date>
    <item>
      <title>Basic Code of Using MKL FFT on MIC</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Basic-Code-of-Using-MKL-FFT-on-MIC/m-p/1019980#M19675</link>
      <description>&lt;PRE class="brush:cpp;"&gt;&amp;nbsp;&lt;/PRE&gt;

&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;In our local cluster we have a bunch of MIC's, which are (almost) never esed. I would like to give it the try, but I have no experience iwth the MKL library or MIC. My programs are very simple and they are based on FFT:&lt;/P&gt;

&lt;P&gt;I have an iterative process in which the smae update procedure is applied to the previous data:&lt;/P&gt;

&lt;PRE class="brush:cpp;"&gt;// initialize data rho[lx]

for(t=0;t&amp;lt;tmax;t++)
{

// first step in real space is local
for(i=0;i&amp;lt;lx;i++)
{
nt(i)=rho&lt;I&gt;*rho&lt;I&gt;*rho&lt;I&gt;;
}
//forward FFT rho-&amp;gt;rhok
//forward FFT nt-&amp;gt; ntk

//update in k space 
for(i=0;i&amp;lt;lx/2+1;i++)
{
newrhok(i)=filter1(i)*rhok(i)+filter2(i)*ntk(i); // complex multiplication
}
//inverse FFT newrhok-&amp;gt;newrho
&lt;/I&gt;&lt;/I&gt;&lt;/I&gt;&lt;/PRE&gt;

&lt;P&gt;So my problem is how to ues MKL FFT's on MIC's and how minimize the transfers between cpu and MIC.&lt;/P&gt;

&lt;P&gt;I hope this is the right forum and many thanks in advance&lt;/P&gt;

&lt;P&gt;Cristian&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 11 Dec 2014 12:06:42 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Basic-Code-of-Using-MKL-FFT-on-MIC/m-p/1019980#M19675</guid>
      <dc:creator>Cristian_Vasile_A_</dc:creator>
      <dc:date>2014-12-11T12:06:42Z</dc:date>
    </item>
    <item>
      <title>Cristian,</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Basic-Code-of-Using-MKL-FFT-on-MIC/m-p/1019981#M19676</link>
      <description>&lt;P&gt;Cristian,&lt;/P&gt;

&lt;P&gt;The best way to to learn how to use MKL for Xeon Phi - to look at the examples which were prepared special for such sort of questions.&lt;/P&gt;

&lt;P&gt;in particular for FFT - you can find such example into&amp;nbsp;"&amp;lt;mkl_root&amp;gt;\examples\mic_offload\dftc\"&amp;nbsp;&lt;/P&gt;

&lt;P&gt;--Gennady&lt;/P&gt;</description>
      <pubDate>Fri, 12 Dec 2014 08:29:44 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Basic-Code-of-Using-MKL-FFT-on-MIC/m-p/1019981#M19676</guid>
      <dc:creator>Gennady_F_Intel</dc:creator>
      <dc:date>2014-12-12T08:29:44Z</dc:date>
    </item>
    <item>
      <title>Quote:Gennady Fedorov (Intel)</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Basic-Code-of-Using-MKL-FFT-on-MIC/m-p/1019982#M19677</link>
      <description>&lt;P&gt;&lt;/P&gt;&lt;BLOCKQUOTE&gt;Gennady Fedorov (Intel) wrote:&lt;BR /&gt;&lt;P&gt;&lt;/P&gt;

&lt;P&gt;Cristian,&lt;/P&gt;

&lt;P&gt;The best way to to learn how to use MKL for Xeon Phi - to look at the examples which were prepared special for such sort of questions.&lt;/P&gt;

&lt;P&gt;in particular for FFT - you can find such example into&amp;nbsp;"&amp;lt;mkl_root&amp;gt;\examples\mic_offload\dftc\"&amp;nbsp;&lt;/P&gt;

&lt;P&gt;--Gennady&lt;/P&gt;

&lt;P&gt;&lt;/P&gt;&lt;/BLOCKQUOTE&gt;&lt;P&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Thank you for your reply. I will check the examples on our cluster.&lt;/P&gt;</description>
      <pubDate>Fri, 12 Dec 2014 14:34:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Basic-Code-of-Using-MKL-FFT-on-MIC/m-p/1019982#M19677</guid>
      <dc:creator>Cristian_Vasile_A_</dc:creator>
      <dc:date>2014-12-12T14:34:00Z</dc:date>
    </item>
  </channel>
</rss>

