List Info

Thread: Retrospective and future direction.




Retrospective and future direction.
user name
2006-04-19 22:33:08
On 4/19/06, Bryan Soto <bryan.a.sotogmail.com> wrote:

> > Native? none 
> > But afaik it uses Threads 'n stuff, where they
supported by Java?
> > I think, now that Yarv is almost finished, maybe I
could test running
> > Nitro with that 
> >
>
> I've been meaning to try that as well. I'm fairly
certain it would
> work since they got rails running on yarv.

Oh boy. I'm curious to see how this flys. Nitro is so much
more
dynamic codewise than Rails, I suspect it won't go as
smoothly. But
there no way to know except to try!

If you do, let me know how it fairs.

Thanks,
T.

_______________________________________________
Nitro-general mailing list
Nitro-generalrubyforge.org
h
ttp://rubyforge.org/mailman/listinfo/nitro-general
Retrospective and future direction.
user name
2006-04-20 04:40:07
On Thursday 20 April 2006 07:33, TRANS wrote:
> On 4/19/06, Bryan Soto <bryan.a.sotogmail.com> wrote:
> > > Native? none 
> > > But afaik it uses Threads 'n stuff, where
they supported by Java?
> > > I think, now that Yarv is almost finished,
maybe I could test running
> > > Nitro with that 
> >
> > I've been meaning to try that as well. I'm
fairly certain it would
> > work since they got rails running on yarv.
>
> Oh boy. I'm curious to see how this flys. Nitro is so
much more
> dynamic codewise than Rails, I suspect it won't go as
smoothly. But
> there no way to know except to try!
>
> If you do, let me know how it fairs.

Well, i tried... but ran into simultanous problems with
rubygems 
(facets)/stdlib and minor issues with syntax-changes...
I think it can be done when we just put a bit more effort
into it, but YARV is 
still not the optimal solution.
However, if we really want it to run on YARV it shouldn't
be a problem was 
just not worth the effort for me, i did that only to explore
a bit how it 
compares performance-wise and found out that it's a speedup
of about 120% for 
most programs.

>
> Thanks,
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-generalrubyforge.org
> h
ttp://rubyforge.org/mailman/listinfo/nitro-general
_______________________________________________
Nitro-general mailing list
Nitro-generalrubyforge.org
h
ttp://rubyforge.org/mailman/listinfo/nitro-general
Retrospective and future direction.
user name
2006-04-20 08:04:34
120% ?  More than twice as fast ?  At version .4 ?  Even if
that is 
actually 20%, that is still promising.  I'll have to keep
tabs on that 
thing.  JRuby is currently slow, but they haven't gone
through a 
performance dev cycle yet.
> On Thursday 20 April 2006 07:33, TRANS wrote:
>   
>> On 4/19/06, Bryan Soto <bryan.a.sotogmail.com> wrote:
>>     
>>>> Native? none 
>>>> But afaik it uses Threads 'n stuff, where
they supported by Java?
>>>> I think, now that Yarv is almost finished,
maybe I could test running
>>>> Nitro with that 
>>>>         
>>> I've been meaning to try that as well. I'm
fairly certain it would
>>> work since they got rails running on yarv.
>>>       
>> Oh boy. I'm curious to see how this flys. Nitro is
so much more
>> dynamic codewise than Rails, I suspect it won't go
as smoothly. But
>> there no way to know except to try!
>>
>> If you do, let me know how it fairs.
>>     
>
> Well, i tried... but ran into simultanous problems with
rubygems 
> (facets)/stdlib and minor issues with syntax-changes...
> I think it can be done when we just put a bit more
effort into it, but YARV is 
> still not the optimal solution.
> However, if we really want it to run on YARV it
shouldn't be a problem was 
> just not worth the effort for me, i did that only to
explore a bit how it 
> compares performance-wise and found out that it's a
speedup of about 120% for 
> most programs.
>
>   
>> Thanks,
>> T.
>>
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-generalrubyforge.org
>> h
ttp://rubyforge.org/mailman/listinfo/nitro-general
>>     
> _______________________________________________
> Nitro-general mailing list
> Nitro-generalrubyforge.org
> h
ttp://rubyforge.org/mailman/listinfo/nitro-general
>
>   

_______________________________________________
Nitro-general mailing list
Nitro-generalrubyforge.org
h
ttp://rubyforge.org/mailman/listinfo/nitro-general
Retrospective and future direction.
user name
2006-04-20 08:58:13
In fact it's always around 100-120% given a runtime of
usual ruby of 10 
seconds it will take 5 seconds in yarv in most cases, and
yes, this is
ruby 2.0.0 (Base: Ruby 1.9.0 2006-02-14) [i686-linux]
YARVCore 0.4.0 Rev: 486 (2006-04-11) [opts: ]

http://ra
fb.net/paste/results/OAhxBH21.html (this might be gone
in a few days)

On Thursday 20 April 2006 17:04, Dylan Bruzenak wrote:
> 120% ?  More than twice as fast ?  At version .4 ? 
Even if that is
> actually 20%, that is still promising.  I'll have to
keep tabs on that
> thing.  JRuby is currently slow, but they haven't gone
through a
> performance dev cycle yet.
>
> > On Thursday 20 April 2006 07:33, TRANS wrote:
> >> On 4/19/06, Bryan Soto <bryan.a.sotogmail.com> wrote:
> >>>> Native? none 
> >>>> But afaik it uses Threads 'n stuff,
where they supported by Java?
> >>>> I think, now that Yarv is almost
finished, maybe I could test running
> >>>> Nitro with that 
> >>>
> >>> I've been meaning to try that as well.
I'm fairly certain it would
> >>> work since they got rails running on yarv.
> >>
> >> Oh boy. I'm curious to see how this flys.
Nitro is so much more
> >> dynamic codewise than Rails, I suspect it
won't go as smoothly. But
> >> there no way to know except to try!
> >>
> >> If you do, let me know how it fairs.
> >
> > Well, i tried... but ran into simultanous problems
with rubygems
> > (facets)/stdlib and minor issues with
syntax-changes...
> > I think it can be done when we just put a bit more
effort into it, but
> > YARV is still not the optimal solution.
> > However, if we really want it to run on YARV it
shouldn't be a problem
> > was just not worth the effort for me, i did that
only to explore a bit
> > how it compares performance-wise and found out
that it's a speedup of
> > about 120% for most programs.
> >
> >> Thanks,
> >> T.
> >>
> >>
_______________________________________________
> >> Nitro-general mailing list
> >> Nitro-generalrubyforge.org
> >> h
ttp://rubyforge.org/mailman/listinfo/nitro-general
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-generalrubyforge.org
> > h
ttp://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-generalrubyforge.org
> h
ttp://rubyforge.org/mailman/listinfo/nitro-general
_______________________________________________
Nitro-general mailing list
Nitro-generalrubyforge.org
h
ttp://rubyforge.org/mailman/listinfo/nitro-general
[1-4]

about | contact  Other archives ( Real Estate discussion Medical topics )