<?xml version="1.0" encoding="UTF-8"?>
<!-- Generated on Fri, 25 May 2012 17:37:57 -0700 -->
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <atom:link href="http://software.intel.com/en-us/articles/tera-scale-computing-a-parallel-path-to-the-future/feed/" rel="self" type="application/rss+xml" />
    <title>Intel Software Network Comments Feed</title>
    <link>http://software.intel.com/en-us/articles/tera-scale-computing-a-parallel-path-to-the-future</link>
    <description></description>
    <language>en-us</language>
    <item>
      <title>By Richard Savill</title>
      <description><![CDATA[ The sheer difficulty comes with the territory. This situation in computing has occured before during the days of the 640kb barrier. Ultimately, the barrier was broken. In the future say, when Microsoft develops a new operating system, probably Glass Windows, forgotten passwords will be a thing of the past because a workstation will be more of an interactive object than a computer, whereby the workstation is somewhat of an extension of the user. It simply will not work without the presence of the user. The computer is modeled after the user. It is a bit scary from our point of view in this early decade, but it will be the condition of the world by then to accept such profiling.

The internet must develop as well - as it has. It will ultimately branch into more exotic protocols that are not easily comprehended by our current thought. I guess the best way to describe it is the supernet, whereby information is 'hogged' by certain superservers. The superservers will have the highest power and actually 'contain' our old internet as we know it today - while serving a higher protocol, both faster and unfortunately more controllable. People will flow to this technology because of the speed of almost instanteous access and the old internet protocols will have to be subscribed to separately - sort of. That is the hard one to envision, but storing human history minute by minute is proving to be untenable for our current system. Therefore, I believe you are right, the Age of Tera is compelled by necessity to come into existence. ]]></description>
      <link>http://software.intel.com/en-us/articles/tera-scale-computing-a-parallel-path-to-the-future/#comment-42778</link>
      <pubDate>Fri, 30 Apr 2010 05:04:05 -0700</pubDate>
      <guid isPermaLink="true">http://software.intel.com/en-us/articles/tera-scale-computing-a-parallel-path-to-the-future/#comment-42778</guid>
    </item>
    <item>
      <title>By Todd Bezenek</title>
      <description><![CDATA[ Justin is one of my heroes.  But, his vision needs either more bandwidth between processors (perhaps provided by photonics) or a new programming paradigm.

I think the future is in algorithms which do not depend on exact results from their neighbors, but rather get "good enough" answers from their neighbors.  If the answers are truly not good enough, we can slow down and correct it.

-Todd ]]></description>
      <link>http://software.intel.com/en-us/articles/tera-scale-computing-a-parallel-path-to-the-future/#comment-51752</link>
      <pubDate>Sat, 13 Nov 2010 00:42:23 -0800</pubDate>
      <guid isPermaLink="true">http://software.intel.com/en-us/articles/tera-scale-computing-a-parallel-path-to-the-future/#comment-51752</guid>
    </item>
  </channel></rss>
