<?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 performance and +DAportable in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546834#M724844</link>
    <description>I have a large application that is written in C and COBOL.  I want to build it on a PA RISC 2.0 chip but I want my customers who still have PA 1 chips to be able to run it so I compile with the +DAportable option.  I want to know if there are any performance implications of this when compared with buildingthe application on a PA 1 chip.</description>
    <pubDate>Fri, 29 Jun 2001 10:26:34 GMT</pubDate>
    <dc:creator>Ian Lochray</dc:creator>
    <dc:date>2001-06-29T10:26:34Z</dc:date>
    <item>
      <title>performance and +DAportable</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546834#M724844</link>
      <description>I have a large application that is written in C and COBOL.  I want to build it on a PA RISC 2.0 chip but I want my customers who still have PA 1 chips to be able to run it so I compile with the +DAportable option.  I want to know if there are any performance implications of this when compared with buildingthe application on a PA 1 chip.</description>
      <pubDate>Fri, 29 Jun 2001 10:26:34 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546834#M724844</guid>
      <dc:creator>Ian Lochray</dc:creator>
      <dc:date>2001-06-29T10:26:34Z</dc:date>
    </item>
    <item>
      <title>Re: performance and +DAportable</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546835#M724845</link>
      <description>&lt;BR /&gt;I have done this kind of thing and didn't notice nor measure any performance implications.&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 29 Jun 2001 10:42:47 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546835#M724845</guid>
      <dc:creator>Rainer von Bongartz</dc:creator>
      <dc:date>2001-06-29T10:42:47Z</dc:date>
    </item>
    <item>
      <title>Re: performance and +DAportable</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546836#M724846</link>
      <description>Hi&lt;BR /&gt;&lt;BR /&gt;Reading information it seems that performances&lt;BR /&gt;are falling... but we do that without measurables&lt;BR /&gt;differences....&lt;BR /&gt;&lt;BR /&gt;Regards&lt;BR /&gt;&lt;BR /&gt;Herv?&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 29 Jun 2001 11:00:41 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546836#M724846</guid>
      <dc:creator>Herve BRANGIER</dc:creator>
      <dc:date>2001-06-29T11:00:41Z</dc:date>
    </item>
    <item>
      <title>Re: performance and +DAportable</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546837#M724847</link>
      <description>Hi Ian,&lt;BR /&gt;&lt;BR /&gt;   I doubt that you will see any degradation in any COBOL application since these tend to be I/O intensive rather that computationally intensive. The only cases in which I have observed any measurable differences in performance involved analysis programs in C involving very large arrays and very complex math. In that case, the performance hit was about 4% (in that particular section of code) but in the vast majority of cases, you will find that the hit is not measurable.&lt;BR /&gt;&lt;BR /&gt;In any event, the proper course is to compile and link using both options and measure but I think all will be well if you choose portable.&lt;BR /&gt;&lt;BR /&gt;Regards, Clay</description>
      <pubDate>Fri, 29 Jun 2001 13:54:21 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546837#M724847</guid>
      <dc:creator>A. Clay Stephenson</dc:creator>
      <dc:date>2001-06-29T13:54:21Z</dc:date>
    </item>
    <item>
      <title>Re: performance and +DAportable</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546838#M724848</link>
      <description>One thing you can do to compromise is to use +DAportable and +DS2.0 to make the compiler use PA1.1 features but schedule instructions to run best on PA2.0 systems.  The default scheduling for +DAportable can be very far from optimal for modern systems.</description>
      <pubDate>Fri, 29 Jun 2001 17:12:31 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546838#M724848</guid>
      <dc:creator>Mike Stroyan</dc:creator>
      <dc:date>2001-06-29T17:12:31Z</dc:date>
    </item>
    <item>
      <title>Re: performance and +DAportable</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546839#M724849</link>
      <description>On our computationally intensive C/C++ application I found at least a 20-30% performance hit on our B1000 when running +DAportable compiled ( latest aCC, HP-UX 10.2).&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Mon, 02 Jul 2001 20:18:40 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-and-daportable/m-p/2546839#M724849</guid>
      <dc:creator>Andrew Cunningham</dc:creator>
      <dc:date>2001-07-02T20:18:40Z</dc:date>
    </item>
  </channel>
</rss>

