cfd9e12239
- Move includes.nav*.sgml to share/sgml/navibar.ent and
<lang>/share/sgml/nabibar.l10n.ent.
- Move includes.sgml and includes.xsl to
share/sgml/common.ent, share/sgml/header.ent, <lang>/share/sgml/l10n.ent,
and <lang>?share/sgml/header.l10n.ent.
- Move most of XSLT libraries to share/sgml/*.xsl and
<lang>/share/sgml/*.xsl.
- Move news.xml and other *.xml files for the similar purpose
to share/sgml/*.xml and <lang>/share/sgml/*.xml.
- Switch to use a custom DTD for HTML document. Now we use
"-//FreeBSD//DTD HTML 4.01 Transitional-Based Extension", which is
HTML 4.01 + some entities previously pulled via
"<!ENTITY % includes SYSTEM "includes.sgml"> %includes;" line.
The location of entity file will be resolved by using catalog file.
- Add DOCTYPE declearation to XML documents. This makes the followings
possible:
* Use of &foo; entities for SGML in an XML file instead of defining
{$foo} as the same content.
* &symbolic; entities for Latin characters.
- Duplicated information between SGML and XML, or English and
translated doc, has been removed as much as possible.
184 lines
4 KiB
Text
184 lines
4 KiB
Text
<!DOCTYPE HTML PUBLIC "-//FreeBSD//DTD HTML 4.01 Transitional-Based Extension//EN" [
|
|
<!ENTITY base CDATA "../..">
|
|
<!ENTITY date "$FreeBSD: www/en/java/docs/performance.sgml,v 1.4 2005/10/04 20:56:33 hrs Exp $">
|
|
<!ENTITY title "FreeBSD &java; Project: Performance Comparison">
|
|
<!ENTITY % navinclude.developers "INCLUDE">
|
|
]>
|
|
<html>
|
|
&header;
|
|
|
|
<h3>Introduction:</h3>
|
|
<p>Much has been talked about recently with Sun's decision to release an
|
|
official port of Java for Linux. This study was done to compare the
|
|
actual performance of the <a href="http://www.blackdown.org">Blackdown</a>
|
|
Linux Port with the <a href="http://www.FreeBSD.org/java/">FreeBSD Port</a>
|
|
of the &jdk;.</p>
|
|
|
|
<h3>Summary:</h3>
|
|
<p>The Linux port is faster than the FreeBSD port, even when the JITs are taken
|
|
into account. For graphic intensive applications, all the FreeBSD tests are
|
|
faster than the Linux, with FreeBSD/TYA being the fastest overall. Comparing
|
|
the Linux port with the FreeBSD port, each with no JIT running, FreeBSD is
|
|
faster. We may conclude that, when running on FreeBSD, the JIT design
|
|
makes the difference in speed. </p>
|
|
|
|
|
|
<h3>Testing Methods:</h3>
|
|
<p>All tests were run on FreeBSD 3.4-RC (Tue Dec 7 09:14:51 EST 1999) on a
|
|
Dual Pentium 200 SMP machine with 128 Megs RAM. Tests were conducted with
|
|
Pendragon Software's <a href="http://www.webfayre.com/pendragon/cm3/index.html">
|
|
Caffeine Mark</a> 3.0 software using the "Run all tests locally" option.</p>
|
|
|
|
<h3>Software Versions:</h3>
|
|
<p>The following is a list of the software used in this test:
|
|
<br>FreeBSD JDK 1.1.8
|
|
<br>Blackdown's JDK 1.2.2-RC3 (under Linux emulation)
|
|
<br><a href="http://www.shudo.net/jit/">ShuJIT</a> 0.3.13
|
|
<br><a href="ftp://gonzalez.cyberus.ca/pub/Linux/java/tya15.tgz">TYA</a> 1.5
|
|
</p>
|
|
|
|
<h3>Summary Results</h3>
|
|
<table class="tblbasic">
|
|
<tr>
|
|
<th></th>
|
|
<th>Test 1</th>
|
|
<th>Test 2</th>
|
|
<th>Test 3</th>
|
|
<th>Test 4</th>
|
|
<th>Test 5</th>
|
|
<th>Average</th>
|
|
</tr>
|
|
<tr>
|
|
<td>Linux JDK 1.2.2/no JIT</td>
|
|
<td>222</td>
|
|
<td>223</td>
|
|
<td>224</td>
|
|
<td>225</td>
|
|
<td>222</td>
|
|
<td>223</td>
|
|
</tr>
|
|
<tr>
|
|
<td>FreeBSD JDK 1.1.8/no JIT</td>
|
|
<td>247</td>
|
|
<td>247</td>
|
|
<td>247</td>
|
|
<td>250</td>
|
|
<td>247</td>
|
|
<td>246</td>
|
|
</tr>
|
|
<tr>
|
|
<td>Linux JDK 1.2.2/standard JIT</td>
|
|
<td>539</td>
|
|
<td>550</td>
|
|
<td>543</td>
|
|
<td>544</td>
|
|
<td>535</td>
|
|
<td>542</td>
|
|
</tr>
|
|
<tr>
|
|
<td>FreeBSD JDK 1.1.8/shujit</td>
|
|
<td>373</td>
|
|
<td>376</td>
|
|
<td>369</td>
|
|
<td>374</td>
|
|
<td>373</td>
|
|
<td>373</td>
|
|
</tr>
|
|
<tr>
|
|
<td>FreeBSD JDK 1.1.8/TYA</td>
|
|
<td>478</td>
|
|
<td>477</td>
|
|
<td>478</td>
|
|
<td>479</td>
|
|
<td>483</td>
|
|
<td>479</td>
|
|
</tr>
|
|
</table>
|
|
|
|
<h3>Sample Detailed Results:</h3>
|
|
<table class="tblbasic">
|
|
<tr>
|
|
<th></th>
|
|
<th>Sieve</th>
|
|
<th>Loop</th>
|
|
<th>Logic</th>
|
|
<th>String</th>
|
|
<th>Float</th>
|
|
<th>Method</th>
|
|
<th>Graphics</th>
|
|
<th>Images</th>
|
|
<th>Dialog</th>
|
|
</tr>
|
|
<tr>
|
|
<td>Linux JDK 1.2.2/no JIT</td>
|
|
<td>380</td>
|
|
<td>434</td>
|
|
<td>433</td>
|
|
<td>489</td>
|
|
<td>363</td>
|
|
<td>337</td>
|
|
<td>306</td>
|
|
<td>25</td>
|
|
<td>40</td>
|
|
</tr>
|
|
<tr>
|
|
<td>FreeBSD JDK 1.1.8/no JIT</td>
|
|
<td>318</td>
|
|
<td>456</td>
|
|
<td>434</td>
|
|
<td>393</td>
|
|
<td>381</td>
|
|
<td>381</td>
|
|
<td>344</td>
|
|
<td>77</td>
|
|
<td>37</td>
|
|
</tr>
|
|
<tr>
|
|
<td>Linux JDK 1.2.2/standard JIT</td>
|
|
<td>1361</td>
|
|
<td>2005</td>
|
|
<td>3354</td>
|
|
<td>872</td>
|
|
<td>1161</td>
|
|
<td>1802</td>
|
|
<td>247</td>
|
|
<td>28</td>
|
|
<td>40</td>
|
|
</tr>
|
|
<tr>
|
|
<td>FreeBSD JDK 1.1.8/shujit</td>
|
|
<td>800</td>
|
|
<td>1566</td>
|
|
<td>2191</td>
|
|
<td>588</td>
|
|
<td>1010</td>
|
|
<td>101</td>
|
|
<td>317</td>
|
|
<td>79</td>
|
|
<td>37</td>
|
|
</tr>
|
|
<tr>
|
|
<td>FreeBSD JDK 1.1.8/TYA</td>
|
|
<td>712</td>
|
|
<td>1467</td>
|
|
<td>2352</td>
|
|
<td>949</td>
|
|
<td>1078</td>
|
|
<td>512</td>
|
|
<td>357</td>
|
|
<td>74</td>
|
|
<td>38</td>
|
|
</tr>
|
|
</table>
|
|
|
|
<h3>Conclusion:</h3>
|
|
<p>Newer versions of the software used in the tests above been released since
|
|
this test was conducted. For this reason, these tests will be run again in
|
|
the near future to see if the newer JITs for FreeBSD alter the results.
|
|
The same JITs will be used with the Linux JDK to test their effects as well.</p>
|
|
|
|
<h3>Tester:</h3>
|
|
<p>These tests were conducted by Patrick Gardella <a href="mailto:patrick@FreeBSD.org">patrick@FreeBSD.org</a> in December 1999.</p>
|
|
&footer;
|
|
</BODY>
|
|
</HTML>
|