Incomplete testing of four XML parsing techniques in Java

XML in peacetime work, will inevitably encounter the XML as a data storage format. In the face of the current variety of solutions, which is the best for us? 　　In this article, I'm going to do an incomplete evaluation of these four mainstream scenarios, just testing for traversing XML, because traversing XML is the most used work (at least I think). Preparation test environment: AMD Poison Dragon 1.4g oc 1.5g, 256m ddr333, WINDOWS2000&NBSP;SERVER&NBSP;SP4, sun jdk 1.4.1+ 　　eclipse 2.1+resin 2.1.8, test in debug mode. 　　The XML file format is as follows: <?xml version= "1.0"  encoding= "GB2312"? > <RESULT> <VALUE> <NO> A1234 </NO> <ADDR> xx County, xx City, Sichuan Province × xx road x section xx </ADDR> </VALUE> <VALUE> <NO> B1234 </NO> <ADDR> Sichuan xx xx Township xx village xx Group </ADDR> </VALUE> test method: Mining Call the bean with the JSP side (as for why the JSP is invoked, refer to: http://blog.csdn.net/rosen/archive/2004/10/15/138324.aspx), so that each solution resolves 10K, 100K, 　　1000K, 10000K XML file that calculates its elapsed time (in milliseconds). JSP file: <%@ page contenttype= "text/html; charset=gb2312"  %><%@ page import= " com.test.* "%>