Anyone has good experience in running PHP in Java/Quercus?
Ryan Chan
to share?
Jerry Stuckle
Why would I want to run a Java implementation and slow my systems to a
crawl? This has to be one of the dumbest idea's I've ever heard of.
--
==================
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
jstu...@attglobal.net
==================
Doug Miller
>Ryan Chan wrote:
>> Anyone use Quercus in the production environment and have any comment
>> to share?
>>
>> http://quercus.caucho.com/
>
>Why would I want to run a Java implementation and slow my systems to a
>crawl? This has to be one of the dumbest idea's I've ever heard of.
>
Jerry Stuckle
ROFLMAO!
Ryan Chan
> Ryan Chan wrote:
>
> Why would I want to run a Java implementation and slow my systems to a
> crawl? This has to be one of the dumbest idea's I've ever heard of.
>
no, it is faster to run PHP in Java, JVM can optimize the PHP code
anyway.
Jerry Stuckle
Java is much slower than C, and a Java interpreter for PHP must be very
slow indeed.
And what makes you think C can't optimize PHP code, also? There are
several accelerators out there.
Now I'm beginning to think you're just a spammer.
Ryan Chan
> Ryan Chan wrote:
> > On Dec 22, 8:24 pm, Jerry Stuckle <jstuck...@attglobal.net> wrote:
> >> Ryan Chan wrote:
>
> >> Why would I want to run a Java implementation and slow my systems to a
> >> crawl? This has to be one of the dumbest idea's I've ever heard of.
>
> > no, it is faster to run PHP in Java, JVM can optimize the PHP code
> > anyway.
>
> Java is much slower than C, and a Java interpreter for PHP must be very
> slow indeed.
>
> And what makes you think C can't optimize PHP code, also? There are
> several accelerators out there.
>
> Now I'm beginning to think you're just a spammer.
>
spammer?
please read this article: http://www.caucho.com/articles/quercus.pdf
I have did similar tests yield similar results.
Jerry Stuckle
Not only is the article full of holes, I wouldn't trust benchmarks from
whomever created the package. Where are independent benchmarks by an
unbiased third party?
For instance:
1. "Each request that travels to Apache gets its own process that is
more or less independent from each other." Not necessarily true.
2. "The processes do not automatically talk with each
other and caching is very limited." No, they don't talk to each other
automatically. But in a web server environment, that is highly
advantageous. And caching does not have to be limited.
3. "Because of Apache's process module, PHP can not take
advantage of the numerous opportunities to cache frequently used
data..." Again, false.
I could continue to shoot holes in the whole article. But what's the point?
r0g
> Ryan Chan wrote:
>> On Dec 22, 11:23 pm, Jerry Stuckle <jstuck...@attglobal.net> wrote:
>>> Ryan Chan wrote:
>>>> On Dec 22, 8:24 pm, Jerry Stuckle <jstuck...@attglobal.net> wrote:
>>>>> Ryan Chan wrote:
>>>>> Why would I want to run a Java implementation and slow my systems to a
>>>>> crawl? This has to be one of the dumbest idea's I've ever heard of.
>>>> no, it is faster to run PHP in Java, JVM can optimize the PHP code
>>>> anyway.
>>> Java is much slower than C, and a Java interpreter for PHP must be very
>>> slow indeed.
>>>
>>> And what makes you think C can't optimize PHP code, also? There are
>>> several accelerators out there.
>>>
>>> Now I'm beginning to think you're just a spammer.
>>>
>>
>> spammer?
>>
>> please read this article: http://www.caucho.com/articles/quercus.pdf
>>
>>
>> I have did similar tests yield similar results.
>
> Not only is the article full of holes, I wouldn't trust benchmarks from
> whomever created the package. Where are independent benchmarks by an
> unbiased third party?
>
Er, he just said he did his own tests. Why don't you do some of your own
before claiming categorically that these speedups are bogus? You ought
to know that theoretical speed means nothing and the only way to know
for sure is through testing.
A fourfold increase in Drupal performance? I'd take such claims with a
little pinch of salt but if I was running a Drupal site across multiple
servers I'd certainly investigate further, it could save a small fortune
if it's even half as good as it claims.
Roger
Jerry Stuckle
Quite frankly, I'm not going to do it because I don't have the time.
However, I can see from the claims on their site that it's bogus.
As for the op - first time posting in this newsgroup. Asks if anyone
has ever used it, as if looking for advice. When challenged, he
immediately starts defending the product he was just asking about.
I small SPAMMER - big time.
r0g
That's a totally natural response though, I don't think it can be taken
as a reliable indicator of someone's core motivations. Being attacked
verbally or textually provokes the same strong threat response from ones
limbic system as the threat of physical attack. People are naturally
going to feel a need to defend themselves when being publicly attacked
with pejoratives like "liar" and "spammer". Of course being new here
(again hardly a cast iron proof of bad intent) he didn't know that he
ought to just ignore you.
Anyway, I took a quick look at their site and there does seem to be more
to their product than the promise of improved performance, although I
would not be so quick to rule that out as you are. If you have to do
tight intergratation with Java systems it might be just what you need:
like Cython when you need to integrate lots of C and Python.
Of course I'm sure this product is useless to YOU but just because you
can't see the point doesn't mean it doesn't exist. Maybe it IS useful to
some others here, I have learned about lots of useful products, modules,
programs and libraries by just seeing them mentioned on usenet.
At the end of the day, if a post has got even a glancing relation to PHP
you ought to let it go, some of it will have value. Responding to real
spammers of the GET CHEEP R0LEX NOW variety never helps anyone and
responding with a tirade of accusations to absolutely everyone who MAY
have a commercial motivation or interest in mentioning a PHP related
product / service is never going to be accurate or fair. In fact I think
it's likely to cause some significant collateral damage to the quality
of the group and the perception of it's community.
Perception of community is a huge factor in many peoples choice between
open platforms. You do want your platform of choice to be popular don't
you? Assuming so, which do you think reflects better on the community...
Uncompromising hostility to commercial interests and suspicion of
newcomers or friendliness and openness towards other users of the platform?
Roger.
Roger.
Jerry Stuckle
I did not attack him verbally or textually until after he started
defending the product he was asking about. That smells like spammer, 100%.
> Anyway, I took a quick look at their site and there does seem to be more
> to their product than the promise of improved performance, although I
> would not be so quick to rule that out as you are. If you have to do
> tight intergratation with Java systems it might be just what you need:
> like Cython when you need to integrate lots of C and Python.
>
I don't believe their performance figures at all. If they are true, it
has to be the worst native PHP implementation on the planet.
Integrate Java with PHP? Yes, there are some places which might need
that - but there are other ways already existing.
> Of course I'm sure this product is useless to YOU but just because you
> can't see the point doesn't mean it doesn't exist. Maybe it IS useful to
> some others here, I have learned about lots of useful products, modules,
> programs and libraries by just seeing them mentioned on usenet.
>
I never said it couldn't be useful to someone else. You, for instance,
very well might bite on it. Looks like you have, in fact.
> At the end of the day, if a post has got even a glancing relation to PHP
> you ought to let it go, some of it will have value. Responding to real
> spammers of the GET CHEEP R0LEX NOW variety never helps anyone and
> responding with a tirade of accusations to absolutely everyone who MAY
> have a commercial motivation or interest in mentioning a PHP related
> product / service is never going to be accurate or fair. In fact I think
> it's likely to cause some significant collateral damage to the quality
> of the group and the perception of it's community.
>
I never said anything about it being off topic here. Just the fact the
op is acting like a spammer.
> Perception of community is a huge factor in many peoples choice between
> open platforms. You do want your platform of choice to be popular don't
> you? Assuming so, which do you think reflects better on the community...
> Uncompromising hostility to commercial interests and suspicion of
> newcomers or friendliness and openness towards other users of the platform?
>
> Roger.
>
That is no excuse to spam usenet or anywhere else. What do you think
this newsgroup would be like if everyone with a product even marginally
related to PHP started posting about it here?