<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Recent changes to feature-requests</title><link>https://sourceforge.net/p/polepos/feature-requests/</link><description>Recent changes to feature-requests</description><atom:link href="https://sourceforge.net/p/polepos/feature-requests/feed.rss" rel="self"/><language>en</language><lastBuildDate>Mon, 19 Oct 2009 06:50:52 -0000</lastBuildDate><atom:link href="https://sourceforge.net/p/polepos/feature-requests/feed.rss" rel="self" type="application/rss+xml"/><item><title>Garbage collection between laps</title><link>https://sourceforge.net/p/polepos/feature-requests/6/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;Garbage Collector should be invoked before each lap, in order to reduce the noise of objects from the previous lap being collected in another lap.&lt;/p&gt;
&lt;p&gt;I placed an invokation to System.gc(); just before watch.start() on line 144 of Circuit class.&lt;/p&gt;
&lt;p&gt;Though GC cannot be forced, this suggest the VM to do its best effort on collecting garbage. I think this invocation is a good idea because othewise I could (in theory) penalize some database because several objects were created for another database.&lt;/p&gt;
&lt;p&gt;One could also argue that GC should be requested between the watch calls.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Javier Godoy</dc:creator><pubDate>Mon, 19 Oct 2009 06:50:52 -0000</pubDate><guid isPermaLink="false">https://sourceforge.neta7930c10e2e5e33b9423a9db8bdadc2336fa1963</guid></item><item><title>Test H2 Database Engine</title><link>https://sourceforge.net/p/polepos/feature-requests/5/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;Hello,&lt;/p&gt;
&lt;p&gt;Is it possible to test the H2 Database Engine as well? The configuration is simple, as it is compatible to HSQLDB. You can download the H2 database here:&lt;/p&gt;
&lt;p&gt;&lt;a href="http://www.h2database.com" rel="nofollow"&gt;http://www.h2database.com&lt;/a&gt;&lt;/p&gt;
&lt;p&gt;Thanks,&lt;br /&gt;
Thomas&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Thomas Mueller</dc:creator><pubDate>Tue, 06 Nov 2007 21:12:01 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net171b8a1b1fc705e67eb4951c55b00d90f268bb22</guid></item><item><title>Update to Hibernate 3</title><link>https://sourceforge.net/p/polepos/feature-requests/4/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;Currently has hibernate 2, should be upgraded to 3.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Travis Reeder</dc:creator><pubDate>Tue, 03 Apr 2007 16:45:16 -0000</pubDate><guid isPermaLink="false">https://sourceforge.neta28abbce7fd5abc240a55ea0ed1210ed4116f880</guid></item><item><title>.NET port</title><link>https://sourceforge.net/p/polepos/feature-requests/3/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;Any hope of porting this project to .NET? &lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">ncoded</dc:creator><pubDate>Thu, 21 Sep 2006 00:56:47 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net7cd5edd713fe2b376cbaf842b93ea81d19389c5f</guid></item><item><title>FOREIGN KEY</title><link>https://sourceforge.net/p/polepos/feature-requests/2/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;Performance tests with foreign key - relations would be &lt;br /&gt;
very interesting, because many real world database &lt;br /&gt;
application make intensively use of foreign keys.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Anonymous</dc:creator><pubDate>Thu, 15 Sep 2005 15:48:20 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net2dba300feaf2364bebfa6ed5551def331ca08e24</guid></item><item><title>Multi-threaded benchmarking environment..</title><link>https://sourceforge.net/p/polepos/feature-requests/1/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;I think pole position is a great tool, it lacks in one&lt;br /&gt;
area however. It would be more usefule of the benchmark&lt;br /&gt;
was really multithreaded so that multiple threads were&lt;br /&gt;
executing concurrently against the database.&lt;/p&gt;
&lt;p&gt;I think that would better mirror a real world&lt;br /&gt;
application as well as better separating the candidate&lt;br /&gt;
DBMS from eachother.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Anonymous</dc:creator><pubDate>Wed, 29 Jun 2005 18:43:03 -0000</pubDate><guid isPermaLink="false">https://sourceforge.netf5a52f2139254561eff013538886e271c5125d9f</guid></item></channel></rss>