|
 Re: Is Groovy the slowest dynamic language?
deamon wrote:
> I've just compared Groovy with other language at
> http://shootout.alioth.debian.org and I am shocked about the really poor
> Groovy performance!
...
>
> Python and Groovy are using similar concepts and Groovy is running on the
> fast JVM. So why ist Groovy so incredible slow?
Is picking out the few performance weaknesses the right way to judge the
overall speed of Groovy?
To me the Groovy performance is absolutely sufficient because of the
easy integration with Java. If something's too slow, I do it in Java.
And Java compared to Python is in most cases much faster.
I appreciate the efforts of the Groovy team to improve the performance,
but if they wouldn't, this would be no real problem to me. Groovy is the
grooviest language with a development team always having the simplicity
and elegance of the language usage in mind - and that counts to me. :-)
Martin
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
Re: Is Groovy the slowest dynamic language?
Right, furthermore, all micro-benchmarks that one comes across over
the internet should really be taken with a grain of salt. If you have
to make a choice, whatever it is, think of the SLA you want to attain
on your application, create a representative test scenario, run it,
and measure. That's the best way to know you're doing the right thing,
by checking by yourself. You'll notice that most of the time, the
performance will be more than enough -- especially if most of the time
is spent talking with a database, or waiting for user input.
On Sun, Feb 24, 2008 at 8:10 PM, Martin Schmidt < martin@...> wrote:
> deamon wrote:
> > I've just compared Groovy with other language at
> > http://shootout.alioth.debian.org and I am shocked about the really poor
> > Groovy performance!
> ...
>
> >
> > Python and Groovy are using similar concepts and Groovy is running on the
> > fast JVM. So why ist Groovy so incredible slow?
>
>
> Is picking out the few performance weaknesses the right way to judge the
> overall speed of Groovy?
>
> To me the Groovy performance is absolutely sufficient because of the
> easy integration with Java. If something's too slow, I do it in Java.
> And Java compared to Python is in most cases much faster.
>
> I appreciate the efforts of the Groovy team to improve the performance,
> but if they wouldn't, this would be no real problem to me. Groovy is the
> grooviest language with a development team always having the simplicity
> and elegance of the language usage in mind - and that counts to me. :-)
>
> Martin
>
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
> http://xircles.codehaus.org/manage_email>
>
>
--
Guillaume Laforge
Groovy Project Manager
G2One, Inc. Vice-President Technology
http://www.g2one.com---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
Re: Is Groovy the slowest dynamic language?
Ditto what others said but also note that the alioth site is still
using Groovy 1.5.0. There was quiet a lot of speed up even in 1.5.4
and the several months of performance work that mostly Alex and also
Jochen have been doing are only in 1.6.
Also, if you search this list you should find the benchmarks I did
showing Groovy faster than Java for the benchmark I chose.
Paul.
deamon wrote:
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
Re: Is Groovy the slowest dynamic language?
On Sun, Feb 24, 2008 at 9:33 PM, deamon < christian.helmbold@...> wrote:
>
> > Is picking out the few performance weaknesses the right way to judge the
> > overall speed of Groovy?
>
> May be the performance in real world applications is not so critical as it
> seems, but the test results show that groovy has fundamental performance
> problems!
Statically typed languages will always be faster than dynamic languages.
> JRuby is also a dynamic language with metaclasses on the jvm and is much
> faster than ruby.
Ruby is a pretty slow language.
Compare JRuby with Java too.
> > To me the Groovy performance is absolutely sufficient because of the
> > easy integration with Java. If something's too slow, I do it in Java.
>
> It sounds a bit like denial of the performance relevance.
Again, microbenchmarks like these are not something I'd rely on, but
I'd work on a plausible scenario and make my measures of my use cases
myself.
--
Guillaume Laforge
Groovy Project Manager
G2One, Inc. Vice-President Technology
http://www.g2one.com---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
Re: Is Groovy the slowest dynamic language?
deamon schrieb:
> It sounds a bit like denial of the performance relevance.
Not really. I'm just solving problems with an elegant language. But I
don't do racing ;-)
Martin
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
Re: Is Groovy the slowest dynamic language?
There is problem but we work on it and each new release is faster
then previous one.
And Guillaume is right that such benchmarks shows almost nothing.
On Mon, Feb 25, 2008 at 12:09 AM, deamon < christian.helmbold@...> wrote:
>
> >> JRuby is also a dynamic language with metaclasses on the jvm and is much
> >> faster than ruby.
>
> I wanted to write "faster than groovy" here.
>
>
> > Ruby is a pretty slow language.
>
> But in these benchmarks Ruby outperforms Groovy.
>
>
> Again, microbenchmarks like these are not something I'd rely on, but
> I'd work on a plausible scenario and make my measures of my use cases
> myself.
>
> To evaluate technlogy for real life use this would be the best way.
> Nevertheless I'm worried about Groovy performance because the differences
> are so dramatic.
>
> Christian
> --
> View this message in context: http://www.nabble.com/Is-Groovy-the-slowest-dynamic-language--tp15667109p15669404.html> Sent from the groovy - user mailing list archive at Nabble.com.
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
> http://xircles.codehaus.org/manage_email>
>
>
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
Re: Is Groovy the slowest dynamic language?
> To me the Groovy performance is absolutely sufficient because of the
> easy integration with Java. If something's too slow, I do it in Java.
It sounds a bit like denial of the performance relevance.
Groovy can be slow and the alioth benchmarks seem to be particularly unfriendly to Groovy. Personally, I wouldn't write compute intensive numerical models in it (Others may beg to differ). However, I use Groovy for writing complex 'buisness' logic against databases. About 90% of the application time is spent in database calls; so basically it doesn't matter how fast the language is. The amount of development time I save for writing in Groovy is worth the performance hit; Groovy is more than fast 'enough' for my application.
I'm not dismissing the fact that Groovy would benefit from optimization. But it's a young language and progressing rapidly. I'm optimistic that performance will get much better in the near future.
-- ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Brian Schlining
|
|
Re: Is Groovy the slowest dynamic language?
With everything I've done using Groovy in production websites, I've
never noticed any performance hits vs Java and Groovy. That being
said, I've used Groovy as a scripting language (meaning non-compiled in
this reference) and compiled classes. With as much traffic as I've had
hitting those sites I have not found any hits on performance. In
addition, I've used Groovy extensively with JME ( jmonkeyengine.com) in
scripting and compiled classes and have yet to notice any differences
in framerates, game play, sound, movement, etc. I don't know anything about the benchmarks between JRuby and Ruby vs Java/Groovy.... whatever. I
can say that compiled classes (according to internal class time prints)
perform the same as the Java ones (which was to be expected) As
for the Groovy scripts (non-compiled), there was an original lag when
the script is first loaded into the jvm, but nothing after that (and
the lag was < 200ms... negligible at best -- unless you compound
that by hundreds of scripts, which is another argument).
Statistics can prove and disprove any point of view. If you want to try, simply take a few Java classes rename them .groovy and run them with your own performance tests, see what happens... if there's anything that's too catastrophic for your tastes, then put the .java back on there and go back... But don't forget to go compiled and non-compiled Groovy scripts just to ensure a well-rounded test.
timo On Sun, Feb 24, 2008 at 3:25 PM, Brian Schlining < bschlining@...> wrote:
> To me the Groovy performance is absolutely sufficient because of the
> easy integration with Java. If something's too slow, I do it in Java.
It sounds a bit like denial of the performance relevance.
Groovy can be slow and the alioth benchmarks seem to be particularly unfriendly to Groovy. Personally, I wouldn't write compute intensive numerical models in it (Others may beg to differ). However, I use Groovy for writing complex 'buisness' logic against databases. About 90% of the application time is spent in database calls; so basically it doesn't matter how fast the language is. The amount of development time I save for writing in Groovy is worth the performance hit; Groovy is more than fast 'enough' for my application.
I'm not dismissing the fact that Groovy would benefit from optimization. But it's a young language and progressing rapidly. I'm optimistic that performance will get much better in the near future.
--
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
|
|
Re: Is Groovy the slowest dynamic language?
On Sun, Feb 24, 2008 at 4:25 PM, Brian Schlining < bschlining@...> wrote:
> ...However, I use Groovy for
> writing complex 'buisness' logic against databases. About 90% of the
> application time is spent in database calls; so basically it doesn't matter
> how fast the language is. The amount of development time I save for
> writing in Groovy is worth the performance hit; Groovy is more than fast
> 'enough' for my application.
>
> I'm not dismissing the fact that Groovy would benefit from optimization. But
> it's a young language and progressing rapidly. I'm optimistic that
> performance will get much better in the near future. --
This is really the key. Nobody will try to assert that Groovy is
blazingly fast, but it gives _the_best_ productivity if you have a
heavy Java foundation and you want the benefit of a dynamic language.
I have never bothered to look at the benchmarks for Groovy; I just
started using it and I have never had a problem with performance for
anything I've used it for. Isn't that what's important?
Not to mention, as many people pointed out, that speed is a priority
for the language and the coming months are going to show great
improvements in those numbers.
-Tom
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
Re: Is Groovy the slowest dynamic language?
On Sun, 2008-02-24 at 10:01 -0800, deamon wrote:
> I've just compared Groovy with other language at
> http://shootout.alioth.debian.org and I am shocked about the really poor
> Groovy performance!
I have just compared the performance of assembly language, C , C++, and
Python. C and C++ are vastly slower than assembler and Python is really
hugely slower than assembler. I am therefore shocked that people
program in C, C++ and Python when assembly language is so much faster.
Everyone must surely program only in assembler or they will not be able
to obtain the correct performance allowed by the processor. C and C++
are clearly poor languages to program in, and Python, Ruby and Groovy
are just the pits when it come to performance so they should be banned
for allowing so much degradation in run time performance.
If anyone is missing the irony and satire then don't.
To be honest I can only treat this as yet another troll. Aren't there
times when you just hate micro-benchmarks. I will not write about my
opinions of people who are controlled by them at the expense of looking
at the overall issues including expressivity, ease of programming,
fitness for context, etc. Clearly I will never use Groovy for complex
QCD or CFD calculations, but then neither would I use assembly language.
Context is everything in this situation and people who rant on about
performance without regard to context are getting close to my threshold
of finding censorship acceptable.
Sure Groovy is not the fastest language in the world. Sure I would like
it to be faster. But actually I don't care that Groovy is 800 times
slower than Java because I use Groovy for what Groovy is good at and
Java for what Java is good at and ALL IN THE SAME PROGRAM. Sorry for
shouting. I'll go to bed now and calm down.
--
Russel.
====================================================
Dr Russel Winder Partner
Concertant LLP t: +44 20 7193 9203
41 Buckmaster Road, f: +44 8700 516 084
London SW11 1EN, UK. m: +44 7770 465 077
|
|
Re: Is Groovy the slowest dynamic language?
> > It sounds a bit like denial of the performance relevance.
>
> Not really. I'm just solving problems with an elegant language. But I
> don't do racing ;-)
I am using using Groovy for 4 weeks now, so you should take my
opinions with a grain of salt (newbie talk).
I do racing ;) and Groovy is the fastest language that I know, at
least in the points that matter to me:
1. Fast learning curve. Oh boy, put fast in that coming from a Java background.
2. Developer time units per functionality developed.
Tiago
Tiago
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
RE: Is Groovy the slowest dynamic language?
Yeah, what HE said.
You know, can we all just agree that Groovy is not the fastest executing language on the planet and get on with it? I mean, sure there are things that can be done to improve the situation, but the WHINING and the CRYING and the TEARING OF CLOTHES... ***NOT HELPING!!!*** It sounds like a bunch of ninnies at a comic-book convention ("Spiderman is cooler than Superman." "Nuh uh!")
For those people who get their emotional satisfaction out of the pure speed of the language they are writing in, may I suggest pure ANSI C (C++ will be too slow if used properly). It is almost as fast as Assembler, portable to different platforms, and prints "Hello World" blazingly fast!
I don't know if this email has helped anyone else, but it certainly has helped me. Thanks for listening.
Jason Smith
-----Original Message-----
From: Tiago Antão [mailto: tiagoantao@...]
Sent: Monday, February 25, 2008 4:45 AM
To: user@...
Subject: Re: [groovy-user] Is Groovy the slowest dynamic language?
> > It sounds a bit like denial of the performance relevance.
>
> Not really. I'm just solving problems with an elegant language. But I
> don't do racing ;-)
I am using using Groovy for 4 weeks now, so you should take my opinions with a grain of salt (newbie talk).
I do racing ;) and Groovy is the fastest language that I know, at least in the points that matter to me:
1. Fast learning curve. Oh boy, put fast in that coming from a Java background.
2. Developer time units per functionality developed.
Tiago
Tiago
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
Re: Is Groovy the slowest dynamic language?
On Mon, Feb 25, 2008 at 3:16 PM, Smith, Jason, CTR, OASD(HA)/TMA
< Jason.Smith.ctr@...> wrote:
> Yeah, what HE said.
>
> You know, can we all just agree that Groovy is not the fastest executing language on the planet and get on with it? I mean, sure there are things that can be done to improve the situation, but the WHINING and the CRYING and the TEARING OF CLOTHES... ***NOT HELPING!!!*** It sounds like a bunch of ninnies at a comic-book convention ("Spiderman is cooler than Superman." "Nuh uh!")
>
> For those people who get their emotional satisfaction out of the pure speed of the language they are writing in, may I suggest pure ANSI C (C++ will be too slow if used properly). It is almost as fast as Assembler, portable to different platforms, and prints "Hello World" blazingly fast!
>
> I don't know if this email has helped anyone else, but it certainly has helped me. Thanks for listening.
At least, it made me laugh :-)
I particularly loved the Spiderman vs Superman argument.
Thanks for that!
--
Guillaume Laforge
Groovy Project Manager
G2One, Inc. Vice-President Technology
http://www.g2one.com---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
RE: Is Groovy the slowest dynamic language?
On Mon, 2008-02-25 at 07:16 -0700, Smith, Jason, CTR, OASD(HA)/TMA
wrote:
> For those people who get their emotional satisfaction out of the pure speed of the language they are writing in, may I suggest pure ANSI C (C++ will be too slow if used properly). It is almost as fast as Assembler, portable to different platforms, and prints "Hello World" blazingly fast!
Not actually true in general though. In general, code generated from
well written C++ source is likely to be smaller and faster than code
generated from well written C source.
--
Russel.
====================================================
Dr Russel Winder Partner
Concertant LLP t: +44 20 7193 9203
41 Buckmaster Road, f: +44 8700 516 084
London SW11 1EN, UK. m: +44 7770 465 077
|
|
Re: Is Groovy the slowest dynamic language?
Scalability concerns are always warranted in this industry but sometimes you just have to consider investment costs first... how many full-time employees at what hourly rate does it take to generate your Important Application with all of the features you want in C or C++?
How many hours of work instead if using Groovy or a dynamic language? Does the savings justify a faster computer for your dev or client production environment (or both)? Devs tend to place an emphasis on writing Beautiful Code when merely functional code would be enough for the clients.
Of course that kind of thinking (Beautiful Code) can help attract some of the more talented (or snooty? or european...) devs who would maybe rather not even talk to you in other case. Of course my opinion take it for what it's worth, I'm only first year out of CS-BS in NY.
Kingdon On Mon, Feb 25, 2008 at 2:57 PM, Russel Winder < russel.winder@...> wrote:
On Mon, 2008-02-25 at 07:16 -0700, Smith, Jason, CTR, OASD(HA)/TMA
wrote:
> For those people who get their emotional satisfaction out of the pure speed of the language they are writing in, may I suggest pure ANSI C (C++ will be too slow if used properly). It is almost as fast as Assembler, portable to different platforms, and prints "Hello World" blazingly fast!
Not actually true in general though. In general, code generated from
well written C++ source is likely to be smaller and faster than code
generated from well written C source.
--
Russel.
====================================================
Dr Russel Winder Partner
Concertant LLP t: +44 20 7193 9203
41 Buckmaster Road, f: +44 8700 516 084
London SW11 1EN, UK. m: +44 7770 465 077
-- -- Kingdon Barrett kingdon.barrett@...
|
|
RE: Is Groovy the slowest dynamic language?
Are you calling me "European?" Or just "snooty?"
K, when you have had to maintain a codebase over a 10 year period where
the code has been written and rewritten by pragmatic neophytes (is that
a snooty word?), you'll realize that beautiful code is a beautiful
thing. But by then it will be too late.
My apologies, and undying thanks, to the wonderful people who happen to
live in Europe (you know who you are) and are actually doing something
to make the code better and faster. :-) And remember, never bait a
troll from NY.
jason.smith.ctr@... (Colorado, just across the pond)
-----Original Message-----
From: Kingdon Barrett [mailto: kpb.west@...]
Sent: Monday, February 25, 2008 1:40 PM
To: user@...
Subject: Re: [groovy-user] Is Groovy the slowest dynamic language?
...
Of course that kind of thinking (Beautiful Code) can help attract some
of the more talented (or snooty? or european...) devs who would maybe
rather not even talk to you in other case. Of course my opinion take it
for what it's worth, I'm only first year out of CS-BS in NY.
Kingdon
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
Re: Is Groovy the slowest dynamic language?
Russel Winder wrote:
> On Mon, 2008-02-25 at 07:16 -0700, Smith, Jason, CTR, OASD(HA)/TMA
> wrote:
>
>> For those people who get their emotional satisfaction out of the pure speed of the language they are writing in, may I suggest pure ANSI C (C++ will be too slow if used properly). It is almost as fast as Assembler, portable to different platforms, and prints "Hello World" blazingly fast!
>
> Not actually true in general though. In general, code generated from
> well written C++ source is likely to be smaller and faster than code
> generated from well written C source.
>
Is this the same Russel who wrote
"I have just compared the performance of assembly language, C , C++, and
Python. C and C++ are vastly slower than assembler and Python is really
hugely slower than assembler."
I know you were baiting a (perceived) troll, but why don't you give us
the numeric equivalent of "vastly slower"?
Justin
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
|
RE: Is Groovy the slowest dynamic language?
Good that I read this mail before answering to Kingdon,
> K, when you have had to maintain a codebase over a 10 year
> period where the code has been written and rewritten by
> pragmatic neophytes (is that a snooty word?), you'll realize
> that beautiful code is a beautiful thing. But by then it
> will be too late.
as I would have written almost exactly the same.
(Only would reduce the period to half the time, in actual
experience of an over 7 MLOC J2EE Project).
Just a hint for you Kingdon: Beauty is not only a matter of
cosmetics or luxury art.
Ask e.g. Steven Weinberg, a famous (non-european) particle
physicist) about the necessity of a scientific theory to
be "beauty".
And see:
http://cogsci.uwaterloo.ca/Articles/beauty.pdfAlso e.g.:
http://everything2.com/index.pl?node_id=99124KR
Det
***********************************************************************
Die Information in dieser email ist vertraulich und ist ausschliesslich
fuer den/die benannten Adressaten bestimmt. Ein Zugriff auf diese
email durch andere Personen als den/die benannten Adressaten ist
nicht gestattet. Sollten Sie nicht der benannte Adressat sein, loeschen
Sie bitte diese email.
***********************************************************************
ARGE Informationssysteme
in der gesetzlichen Krankenversicherung GmbH
Paul-Klinger-Strasse 15, 45127 Essen
Amtsgericht Essen HRB 11122
Vorsitzender des Aufsichtsrates: Wolfgang Schmeinck
Geschaeftsfuehrer: Gernot Kiefer
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
|
signature.asc (196 bytes) 