[Mono-dev] performance and stress test

[Jonathan Chambers]

Performance is an area I am interested in. I'd be willing to do what Miguel
describes, but I'd like some direction.

I assume I could set up a series of parallel mono environments based on
previous releases. Then, I could run test sets on each release and see how
we are improving (or getting worse). I'd then run the sets on any new
release and on svn from time to time. I could do that on Windows and on
Linux.

So, does this sound valid? And if so, what type of tests should we run? I do
have the book '*Maximizing .NET Performance'* (
http://www.amazon.com/Maximizing-NET-Performance-Experts-Voice/dp/1590591410),
and could run some of the sample performance tests from it. Also, the tests
from the language shootout link above. Any other thoughts/directions?



Thanks,
Jonathan


On 4/28/07, Miguel de Icaza <miguel at novell.com> wrote:
>
> Hello,
>
> > I found thi site :
> > http://shootout.alioth.debian.org/gp4/benchmark.php?test=all
> 〈=csharp&lang2=java
> >
> > which is good to see a global test and not a detailed test.
>
> Yes, we see that one frequently.
>
> It has a couple of problems though, it compares oranges to apples.   One
> of the tests for example compares threading primitives in Mono vs yield
> in python.
>
> The other problem is that this tests the released versions of compilers,
> and what would be interesting is to track the progress that we are
> making and also track how this compares against .NET as comparing to
> other systems (and specially different implementations) introduces too
> many unrelated variables.
>
> > bye
> > duff
> >
> > Selon Miguel de Icaza <miguel at novell.com>:
> >
> > > Hello,
> > >
> > > > Is there an tool or a batch on svn to have a report about
> performance of
> > > mono
> > > > Versus Ms.NET implementation.
> > > >
> > > > I know that there is the class status script. I search something
> quite
> > > simple to
> > > > be run as class status to get out performance index compared to ms
> > > performance
> > > > index.
> > >
> > > You might want to track the progress that Mono makes, this is being
> > > tracked by the guys at Charles University in Prague, they have done a
> > > phenomenal job:
> > >
> > >     http://dsrg.mff.cuni.cz/projects/mono/index.phtml
> > >
> > > from the project:
> > >
> > >                 "The purpose of this project is detection of
> performance
> > >                 regressions in daily development versions of MONO.
> > >                 Selected benchmarks are compiled and run for each
> daily
> > >                 version. The daily versions are compiled MONO daily
> > >                 packages. Selected plots from the benchmarks follow.
> > >                 Description of the system used for benchmarking is
> here.
> > >                 The benchmarks are run on a dedicated machine, with
> > >                 disabled network interface and with no system services
> > >                 running."
> > >
> > > Am not aware of someone routinely adding new benchmarks and publishing
> > > the results of performance as time goes by between Mono and .NET.   It
> > > would be interesting to have.
> > >
> >
> >
> _______________________________________________
> Mono-devel-list mailing list
> Mono-devel-list at lists.ximian.com
> http://lists.ximian.com/mailman/listinfo/mono-devel-list
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ximian.com/pipermail/mono-devel-list/attachments/20070428/355ec837/attachment.html