<?xml version="1.0" encoding="UTF-8"?>
<!-- Generated on Fri, 25 May 2012 17:57:35 -0700 -->
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <atom:link href="http://software.intel.com/en-us/articles/the-three-stages-of-preparation-for-optimizing-parallel-software/feed/" rel="self" type="application/rss+xml" />
    <title>Intel Software Network Comments Feed</title>
    <link>http://software.intel.com/en-us/articles/the-three-stages-of-preparation-for-optimizing-parallel-software</link>
    <description></description>
    <language>en-us</language>
    <item>
      <title>By jhswartz3141</title>
      <description><![CDATA[ I like this article.  The basics are very correct.  The title is what threw me off.

Optimizing parallel software requires a first step that was not mentioned in this article: Optimizing Single CPU Software.  The methodology listed here in a clear concise way applies equally to optimizing single CPU software and after the single CPU software part of the parallel software is optimized, you can be sure that the real life work loads that you are modeling for parallel optimization have the best possible single CPU run times.

Thank you for the article! ]]></description>
      <link>http://software.intel.com/en-us/articles/the-three-stages-of-preparation-for-optimizing-parallel-software/#comment-31635</link>
      <pubDate>Sat, 26 Sep 2009 12:10:17 -0700</pubDate>
      <guid isPermaLink="true">http://software.intel.com/en-us/articles/the-three-stages-of-preparation-for-optimizing-parallel-software/#comment-31635</guid>
    </item>
    <item>
      <title>By thiamchunkoh</title>
      <description><![CDATA[ The methodology is an interative Software Engineering approach based on MPI and middleware on message oriented Middleware based on the latest technology on CORBA, DCOM, COM, Active X on the different platform on Java EE or Microsoft. Net Framework. the Architecture of the whole software engineeing metholodgies is based on data driven architecture and native memory allocation on the system stack ot the heaps. With code profiling with Linear and parallel programming architectures based on Model Driven Architecture (PIM or PSM) I think it will benefit the High Performance Computing Community Individuals for this methologies implemented above as required in Software Configuration Management and versioning ]]></description>
      <link>http://software.intel.com/en-us/articles/the-three-stages-of-preparation-for-optimizing-parallel-software/#comment-37011</link>
      <pubDate>Fri, 18 Dec 2009 22:57:26 -0800</pubDate>
      <guid isPermaLink="true">http://software.intel.com/en-us/articles/the-three-stages-of-preparation-for-optimizing-parallel-software/#comment-37011</guid>
    </item>
    <item>
      <title>By New to Parallel Programming? &amp;#8211; Intel Software Network Blogs - Intel&amp;#174; Software Network</title>
      <description><![CDATA[ n/a ]]></description>
      <link>http://software.intel.com/en-us/articles/the-three-stages-of-preparation-for-optimizing-parallel-software/#comment-61332</link>
      <pubDate>Thu, 30 Jun 2011 11:40:02 -0700</pubDate>
      <guid isPermaLink="true">http://software.intel.com/en-us/articles/the-three-stages-of-preparation-for-optimizing-parallel-software/#comment-61332</guid>
    </item>
  </channel></rss>
