XOM API for Processing XML with Java

[Wolfgang Hoschek <whoschek AT lbl.gov>]

On Dec 5, 2004, at 2:19 AM, Elliotte Harold wrote:

> It strikes me that reporting performance issues is similar to 
> reporting bugs. A problem will get addressed sooner if it comes with a 
> small, self-contained, reproducible test case, rather than simply a 
> report. A report that doesn't provide a test case will still be 
> addressed, but later rather than sooner because then I have to invent 
> a test case that exposes the issue.
>
> This is even more true in the case of optimization than bug reporting 
> because it's relatively hard to write the test case, and many tests 
> prove not to be reproducible across different hardware and VMs. Some 
> bugs are obvious. Performance issues almost never are.

I've mentioned before the problems involved with generating and 
interpreting meaningful benchmarks, in particular microbenchmakrs. As 
always, my tests use the binary codec, which is a meaningful way to 
measure the extent of any real improvements in this area.