One issue will be repeatability. We should prejudice our benchmark
tracking based on configurations using project facilities, like build-
test.
geir
On Jan 26, 2007, at 11:58 AM, Rana Dasgupta wrote:
That sounds like the right way. We will, put up a page on Harmony
with
results on some typical benchmarks like Scimark, DaCapo including
configuration used, results etc. We can use that as a start to
understand
progress over time.
If folks are interested in tracking more benchmarks, they can add
to the
page and post results that they are actively tracking.
Rana
On 1/25/07, Alexey Petrenko <[EMAIL PROTECTED]> wrote:
2007/1/26, Robin Garner <[EMAIL PROTECTED]>:
> Two suggestions:
>
> 1) Control (and report!) the heap size (using -Xmx and -Xms).
There is a lot of VM options which affects performance and we
definetly plan to report full config: hardware/software configuration
and list of options.
> 2) Use '-n 3' and report all 3 iterations.
>
> cheers
>
> > We are planning to start running benchmarks regularly on
Harmony in
> > the nearest future. We are testing the infrastructure now...
> >
> > SY, Alexey
> >
> > 2007/1/25, Leo Li <[EMAIL PROTECTED]>:
> >> Hi, all:
> >> With the progress of Harmony project, I think,
performance is
> >> gradually
> >> becoming an important feature for our real customers. Besides
the
> >> stability
> >> and compatibility, users would incline to choose the one with
higher
> >> speed. Aside from VM, classlib itself also plays a
significant role
in
> >> performance.
> >> Although we now know some issues that are performance
related, in
a
> >> long
> >> run, we need performace test results to pick out the hotspot in
Harmony
> >> project. Without data, we are like blind men running in the
dark, in
the
> >> area of performance.
> >> Is there any advice about the tools to test the
performance of
> >> Harmony?
> >> Actually, if my remember is right, some guys have actually
run the
> >> performance tests. (But I forget where the mail is.)Maybe we
can post
> >> the
> >> result on Harmony wiki thus we can have a baseline to improve
our
> >> code.:)
> >>
> >> Thanks.
> >> --
> >> Leo Li
> >> China Software Development Lab, IBM
> >>
> >>
> >
>
>
>
