Comments on: Performance tests for introspection of JavaBeans http://www.christianschenk.org/blog/performance-tests-for-introspection-of-javabeans/ Writing about my experiences with technology and all different kinds of projects and experiments Thu, 04 Dec 2008 03:52:55 +0000 By: Christian Schenk http://www.christianschenk.org/blog/performance-tests-for-introspection-of-javabeans/#comment-465 Christian Schenk Sun, 02 Dec 2007 15:49:44 +0000 http://www.christianschenk.org/blog/performance-tests-for-introspection-of-javabeans/#comment-465 As a follow up to my tests with the Introspector from the Java API, I did some more testing with various Java Bean to Java Bean mappers. Have a look at it <a href="http://www.christianschenk.org/blog/java-bean-mapper-performance-tests/" rel="nofollow">here</a>. As a follow up to my tests with the Introspector from the Java API, I did some more testing with various Java Bean to Java Bean mappers. Have a look at it here.

]]> By: Christian Schenk http://www.christianschenk.org/blog/performance-tests-for-introspection-of-javabeans/#comment-94 Christian Schenk Sat, 11 Aug 2007 16:03:02 +0000 http://www.christianschenk.org/blog/performance-tests-for-introspection-of-javabeans/#comment-94 Hi Simon, thank you for the link to <a title="Dozer" href="http://dozer.sourceforge.net">Dozer</a>. I'll check it out and write a (hopefully detailed) post soon. I obviously forgot to upload the test code: now you can have a look at it <a href="http://data.christianschenk.org/performance-tests-for-introspection-of-javabeans/xref/org/christianschenk/beanintrospect/PerformanceComparison.html">here</a>. Basically it's a "big for loop" and certainly not the most realistic way to test this, but I think that it gives us a good indication for the one or the other solution - whether this might be faster than that and so on. Do you have another idea for a performance test - instead of the "big for loop"? Hi Simon,

thank you for the link to Dozer. I’ll check it out and write a (hopefully detailed) post soon.

I obviously forgot to upload the test code: now you can have a look at it here. Basically it’s a “big for loop” and certainly not the most realistic way to test this, but I think that it gives us a good indication for the one or the other solution - whether this might be faster than that and so on.

Do you have another idea for a performance test - instead of the “big for loop”?

]]> By: Simon Lebettre http://www.christianschenk.org/blog/performance-tests-for-introspection-of-javabeans/#comment-93 Simon Lebettre Fri, 10 Aug 2007 15:45:50 +0000 http://www.christianschenk.org/blog/performance-tests-for-introspection-of-javabeans/#comment-93 Hi, facing exactly the same problem (having to write an enormous load of boring code to map from JAXB objects to business DTOs), I found this framework : http://dozer.sourceforge.net In my own performance tests (real use on the server with logs), i see the mapping (after the first time it encounters a new object wich caches all the methods lookups) is taking 0 ms , whenever I use hand coded mapping or dozer. By the way , I figured that for this matter, testing in a big for loop on my dev computer makes no sense, I think I end up testing what my operating system is doing! - Could you try your perf test with dozer ? - Could you explain your performance tests methodology / provide the test code ? Hi, facing exactly the same problem (having to write an enormous load of boring code to map from JAXB objects to business DTOs), I found this framework : http://dozer.sourceforge.net

In my own performance tests (real use on the server with logs), i see the mapping (after the first time it encounters a new object wich caches all the methods lookups) is taking 0 ms , whenever I use hand coded mapping or dozer.

By the way , I figured that for this matter, testing in a big for loop on my dev computer makes no sense, I think I end up testing what my operating system is doing!

- Could you try your perf test with dozer ?
- Could you explain your performance tests methodology / provide the test code ?

]]>