Comment #5 on issue 23 by brainslog: Diameter Stack: Improve Diameter
Router (Performance & Footprint)
http://code.google.com/p/jdiameter/issues/detail?id=23
Hi Paul,
Thanks for your answers. Here's some more to add to discussion:
1) Regarding collisions, they may happen.. usually in a test scenario, not
as much is a real-world scenario. While the string concatenation is heavy,
related to using a long, it may not impact as much in the overall process.
Also the hash process is heavier for String. But, I've ran some performance
checks and it seems to behave fairly well.
2) Agree on this too, and after some performance tests I can confirm the
change is for the better, and will scale better.
3) Right, as mentioned, this should not happen... only if there are leaks,
so it's fairly safe to have it.
Thanks for these changes, we will integrate them on our code.
