<?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 Finally my new scalability mathematical model is here... in Intel® Moderncode for Parallel Architectures</title>
    <link>https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Finally-my-new-scalability-mathematical-model-is-here/m-p/817460#M1186</link>
    <description>&lt;P&gt;&lt;BR /&gt;Hello,&lt;/P&gt;&lt;P&gt;Now to generalize my mathematical model and make it more clear,&lt;BR /&gt;if we are running n threads in parallel in a time interval and entering&lt;BR /&gt;n critical sections in the same time interval, my mathematical model&lt;BR /&gt;will be equal to:&lt;/P&gt;&lt;P&gt;Speed(t,N,,Sic,Soc) = 1 / (A + (P / N)) &lt;/P&gt;&lt;P&gt;A = ((Sic_1 * ((_1-1) * (_1)) / (2 * _1)) + (Sic_2 * ((_2-1) &lt;BR /&gt; * (_2)) / (2 * _2)) + .... (Sic_n * ((_n-1) * (_n)) / (2 * _n)))/N&lt;/P&gt;&lt;P&gt;A = (Sum (i: 1 -&amp;gt;n ) (Sic_n * ((_n-1) * (_n)) / (2 * _n)))) / N&lt;/P&gt;&lt;P&gt;Sic_1...Sic_n: serial parts inside the critical regions(locked regions)&lt;BR /&gt;P: parallel part&lt;BR /&gt;n: number of threads (or processes)&lt;BR /&gt;N: number of cores&lt;BR /&gt;C_1...C_n: contention factors in percentage for each crtitical region&lt;/P&gt;&lt;P&gt;C = Number of contended threads (in the critical region) / n&lt;BR /&gt;and  = C * n &lt;/P&gt;&lt;P&gt;I wrote in my other article: &lt;/P&gt;&lt;P&gt;&lt;A href="http://pages.videotron.com/aminer/efficiency_availibility.htm"&gt;http://pages.videotron.com/aminer/efficiency_availibility.htm&lt;/A&gt; &lt;/P&gt;&lt;P&gt;"So, as you have noticed you can simulate our enterprise &lt;BR /&gt;webserver with this mathematical analytical equation and &lt;BR /&gt;validate the data collected by fwptt stress webserrer tool &lt;BR /&gt;and try to keep T( the average response time of your HTTP requests) &lt;BR /&gt;bellow or equal to 5 seconds, and see also if your webserver &lt;BR /&gt;is scaling, and we call this also Quality control using OR &lt;BR /&gt;(operational research). " &lt;/P&gt;&lt;P&gt;So, if you want to keep the response time T 'low' in enterprise &lt;BR /&gt;websites, you have to reduce the logical and physical contention &lt;BR /&gt;can occur by reducing the Sic_1...Sic_n (i already explained how to do it)&lt;BR /&gt;and you must also control the  by limiting for example the 'number' &lt;BR /&gt;of requests that can run at the same time and also scale out your&lt;BR /&gt;website etc...&lt;/P&gt;&lt;P&gt;My general scalability mathematical model is also applicable to physical&lt;BR /&gt;contention and logical contention.&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;Read more: &lt;/P&gt;&lt;P&gt;&lt;A href="http://pages.videotron.com/aminer/scalability.htm"&gt;http://pages.videotron.com/aminer/scalability.htm&lt;/A&gt; &lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Regards, &lt;BR /&gt;Amine Moulay Ramdane. &lt;BR /&gt;&lt;A href="http://pages.videotron.com/aminer/"&gt;http://pages.videotron.com/aminer/&lt;/A&gt; &lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;</description>
    <pubDate>Sat, 25 Sep 2010 15:45:01 GMT</pubDate>
    <dc:creator>aminer10</dc:creator>
    <dc:date>2010-09-25T15:45:01Z</dc:date>
    <item>
      <title>Finally my new scalability mathematical model is here...</title>
      <link>https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Finally-my-new-scalability-mathematical-model-is-here/m-p/817459#M1185</link>
      <description>&lt;BR /&gt;Hello,&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Ihave thought all the night, and finallymy new and general scalability &lt;BR /&gt;mathematical model, BETTER than Amadahl's law, is here:&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Read my article, welcome&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://pages.videotron.com/aminer/scalability.htm"&gt;http://pages.videotron.com/aminer/scalability.htm&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;Amine Moulay Ramdane.&lt;BR /&gt;&lt;A href="http://pages.videotron.com/aminer/"&gt;http://pages.videotron.com/aminer/&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Sat, 25 Sep 2010 11:38:39 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Finally-my-new-scalability-mathematical-model-is-here/m-p/817459#M1185</guid>
      <dc:creator>aminer10</dc:creator>
      <dc:date>2010-09-25T11:38:39Z</dc:date>
    </item>
    <item>
      <title>Finally my new scalability mathematical model is here...</title>
      <link>https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Finally-my-new-scalability-mathematical-model-is-here/m-p/817460#M1186</link>
      <description>&lt;P&gt;&lt;BR /&gt;Hello,&lt;/P&gt;&lt;P&gt;Now to generalize my mathematical model and make it more clear,&lt;BR /&gt;if we are running n threads in parallel in a time interval and entering&lt;BR /&gt;n critical sections in the same time interval, my mathematical model&lt;BR /&gt;will be equal to:&lt;/P&gt;&lt;P&gt;Speed(t,N,,Sic,Soc) = 1 / (A + (P / N)) &lt;/P&gt;&lt;P&gt;A = ((Sic_1 * ((_1-1) * (_1)) / (2 * _1)) + (Sic_2 * ((_2-1) &lt;BR /&gt; * (_2)) / (2 * _2)) + .... (Sic_n * ((_n-1) * (_n)) / (2 * _n)))/N&lt;/P&gt;&lt;P&gt;A = (Sum (i: 1 -&amp;gt;n ) (Sic_n * ((_n-1) * (_n)) / (2 * _n)))) / N&lt;/P&gt;&lt;P&gt;Sic_1...Sic_n: serial parts inside the critical regions(locked regions)&lt;BR /&gt;P: parallel part&lt;BR /&gt;n: number of threads (or processes)&lt;BR /&gt;N: number of cores&lt;BR /&gt;C_1...C_n: contention factors in percentage for each crtitical region&lt;/P&gt;&lt;P&gt;C = Number of contended threads (in the critical region) / n&lt;BR /&gt;and  = C * n &lt;/P&gt;&lt;P&gt;I wrote in my other article: &lt;/P&gt;&lt;P&gt;&lt;A href="http://pages.videotron.com/aminer/efficiency_availibility.htm"&gt;http://pages.videotron.com/aminer/efficiency_availibility.htm&lt;/A&gt; &lt;/P&gt;&lt;P&gt;"So, as you have noticed you can simulate our enterprise &lt;BR /&gt;webserver with this mathematical analytical equation and &lt;BR /&gt;validate the data collected by fwptt stress webserrer tool &lt;BR /&gt;and try to keep T( the average response time of your HTTP requests) &lt;BR /&gt;bellow or equal to 5 seconds, and see also if your webserver &lt;BR /&gt;is scaling, and we call this also Quality control using OR &lt;BR /&gt;(operational research). " &lt;/P&gt;&lt;P&gt;So, if you want to keep the response time T 'low' in enterprise &lt;BR /&gt;websites, you have to reduce the logical and physical contention &lt;BR /&gt;can occur by reducing the Sic_1...Sic_n (i already explained how to do it)&lt;BR /&gt;and you must also control the  by limiting for example the 'number' &lt;BR /&gt;of requests that can run at the same time and also scale out your&lt;BR /&gt;website etc...&lt;/P&gt;&lt;P&gt;My general scalability mathematical model is also applicable to physical&lt;BR /&gt;contention and logical contention.&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;Read more: &lt;/P&gt;&lt;P&gt;&lt;A href="http://pages.videotron.com/aminer/scalability.htm"&gt;http://pages.videotron.com/aminer/scalability.htm&lt;/A&gt; &lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Regards, &lt;BR /&gt;Amine Moulay Ramdane. &lt;BR /&gt;&lt;A href="http://pages.videotron.com/aminer/"&gt;http://pages.videotron.com/aminer/&lt;/A&gt; &lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;</description>
      <pubDate>Sat, 25 Sep 2010 15:45:01 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Finally-my-new-scalability-mathematical-model-is-here/m-p/817460#M1186</guid>
      <dc:creator>aminer10</dc:creator>
      <dc:date>2010-09-25T15:45:01Z</dc:date>
    </item>
  </channel>
</rss>

