Excellent!
As soon as I get a chance I will take a look.
I am most interested in doing a run with -XX:+PrintCompilation and then
observing if anything is getting JIT compiled after the warmup phase.
That would give us the best and most accurate metrics of identifying
improvements and/or regressions.
Great work!
charlie ...
Oleksiy Stashok wrote:
> Hello Charlie,
>
> there is warmup phase for each testcase, which is not counted in the
> final result.
>
> WBR,
> Alexey.
>
> charlie hunt wrote:
>> Alexey,
>>
>> I haven't looked at the source code you committed and at japex :-/
>>
>> How does does grizzlypex handle things like ignoring JVM warm up
>> being recorded as part of what is measured ?
>>
>> thanks,
>>
>> charlie ...
>>
>> Oleksiy Stashok wrote:
>>> Hello.
>>> I've commited GrizzlyPex under Grizzly 1.5 modules/grizzlypex.
>>> It's benchmark framework for Grizzly, built on top of Japex [1].
>>> For now it is working with 'ant' and has just 2 testcases to measure
>>> echo speed (request/reply) for different protocols, and controller
>>> start/stop speed. (more info is in readme file).
>>> Current GrizzlyPex version measures bm of Grizzly framework, which
>>> framework.jar file is put to grizzlypex/lib directory.
>>> GrizzlyPex framework will be extended and put on Hudson, where I
>>> will schedule daily bm measurement work.
>>>
>>> WBR,
>>> Alexey.
>>>
>>> [1] japex.dev.java.net
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe_at_grizzly.dev.java.net
>>> For additional commands, e-mail: dev-help_at_grizzly.dev.java.net
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe_at_grizzly.dev.java.net
>> For additional commands, e-mail: dev-help_at_grizzly.dev.java.net
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe_at_grizzly.dev.java.net
> For additional commands, e-mail: dev-help_at_grizzly.dev.java.net
>
--
Charlie Hunt
Java Performance Engineer
630.285.7708 x47708 (Internal)
<http://java.sun.com/docs/performance/>