You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@tapestry.apache.org by Radim Burget <Ra...@seznam.cz> on 2005/09/06 23:29:21 UTC
performance question
It is probably realy stupid question but I can't help.....
I am quite new to java and java related stafs. As I read Java is in
general assumed to be high-performanance against technologies like PHP.
Now I am trying to write simple application in Tapestry with "List of
items" and "details"pages. I am surprised that clicking on each link
takes second or more (PIV 26000MHz, RAM 512MB) (true - with
recompilation forced)
In PHP similar think responses almost immediately....
Is the performance difference affected only by recompilation? It seems
to me to be PHP faster even if I turn forcing of recompilation off....
Thanks in advance.
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
RE: performance question
Posted by Karthik Abram <ka...@neovera.com>.
Have you considered writing your app in assembly? It is faster than PHP or
Java. Its also very compact and esoteric.
-----Original Message-----
From: Radim Burget [mailto:RadimBurget@seznam.cz]
Sent: Tuesday, September 06, 2005 5:29 PM
To: Tapestry users
Subject: performance question
It is probably realy stupid question but I can't help.....
I am quite new to java and java related stafs. As I read Java is in
general assumed to be high-performanance against technologies like PHP.
Now I am trying to write simple application in Tapestry with "List of
items" and "details"pages. I am surprised that clicking on each link
takes second or more (PIV 26000MHz, RAM 512MB) (true - with
recompilation forced)
In PHP similar think responses almost immediately....
Is the performance difference affected only by recompilation? It seems
to me to be PHP faster even if I turn forcing of recompilation off....
Thanks in advance.
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
Re: performance question
Posted by Paul Cantrell <ca...@pobox.com>.
I'm not sure that's entirely fair -- PHP will outstrip Java on the
first request for sure, but a Tapestry + OR app with some decent
caching should be able to outshine PHP by a factor of 2x or more,
even on individual requests.
Pat is completely correct, though: it's about the right tool for the
job, and Java takes much more up-front investment in infrastructure
in exchange the benefits of scalability and maintainability.
Cheers,
Paul
On Sep 6, 2005, at 4:37 PM, Patrick Casey wrote:
>
> Java servlets are *not* going to be faster than PHP, fastCGI, or
> probably even an interpreted perl or python script at handing a single
> request. There's a huge amount of overhead involved in setting up and
> maintaining a servlet context and pool of handler threads, only to
> spin one
> up and bind it with every inbound request.
>
> Where servlets shine is in high volume or other *very* large sites
> where the improved scalability of the platform (like its ability to
> function
> across a cluster, unlike PHP) makes up for the overhead. Likewise,
> it's my
> experience that maintaining and debugging java, especially in a nice
> debugger like eclipse is orders of magnitude easier than debugging
> perl or
> PHP.
>
> The point is, choose the right tool for the right job. Java
> servlets
> are *not* the ideal solution for a quick and dirty crud app for your
> neighbors to post pictures of their houses. That'd be like rolling
> in an
> AS-400 to keep track of your contact list.
>
> --- Pat
>
>
>> -----Original Message-----
>> From: Radim Burget [mailto:RadimBurget@seznam.cz]
>> Sent: Tuesday, September 06, 2005 2:29 PM
>> To: Tapestry users
>> Subject: performance question
>>
>> It is probably realy stupid question but I can't help.....
>>
>> I am quite new to java and java related stafs. As I read Java is in
>> general assumed to be high-performanance against technologies like
>> PHP.
>>
>> Now I am trying to write simple application in Tapestry with "List of
>> items" and "details"pages. I am surprised that clicking on each link
>> takes second or more (PIV 26000MHz, RAM 512MB) (true - with
>> recompilation forced)
>>
>> In PHP similar think responses almost immediately....
>>
>> Is the performance difference affected only by recompilation? It
>> seems
>> to me to be PHP faster even if I turn forcing of recompilation
>> off....
>>
>> Thanks in advance.
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
>> For additional commands, e-mail: tapestry-user-
>> help@jakarta.apache.org
>>
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
RE: performance question
Posted by Karthik Abram <ka...@neovera.com>.
Grrr... I have to cancel that order for Blue Gene/L that I was planning to
run my Tapestry based contact list!
-----Original Message-----
From: Patrick Casey [mailto:patcasey@adelphia.net]
Sent: Tuesday, September 06, 2005 5:38 PM
To: 'Tapestry users'
Subject: RE: performance question
Java servlets are *not* going to be faster than PHP, fastCGI, or
probably even an interpreted perl or python script at handing a single
request. There's a huge amount of overhead involved in setting up and
maintaining a servlet context and pool of handler threads, only to spin one
up and bind it with every inbound request.
Where servlets shine is in high volume or other *very* large sites
where the improved scalability of the platform (like its ability to function
across a cluster, unlike PHP) makes up for the overhead. Likewise, it's my
experience that maintaining and debugging java, especially in a nice
debugger like eclipse is orders of magnitude easier than debugging perl or
PHP.
The point is, choose the right tool for the right job. Java servlets
are *not* the ideal solution for a quick and dirty crud app for your
neighbors to post pictures of their houses. That'd be like rolling in an
AS-400 to keep track of your contact list.
--- Pat
> -----Original Message-----
> From: Radim Burget [mailto:RadimBurget@seznam.cz]
> Sent: Tuesday, September 06, 2005 2:29 PM
> To: Tapestry users
> Subject: performance question
>
> It is probably realy stupid question but I can't help.....
>
> I am quite new to java and java related stafs. As I read Java is in
> general assumed to be high-performanance against technologies like PHP.
>
> Now I am trying to write simple application in Tapestry with "List of
> items" and "details"pages. I am surprised that clicking on each link
> takes second or more (PIV 26000MHz, RAM 512MB) (true - with
> recompilation forced)
>
> In PHP similar think responses almost immediately....
>
> Is the performance difference affected only by recompilation? It seems
> to me to be PHP faster even if I turn forcing of recompilation off....
>
> Thanks in advance.
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
RE: performance question
Posted by Patrick Casey <pa...@adelphia.net>.
Java servlets are *not* going to be faster than PHP, fastCGI, or
probably even an interpreted perl or python script at handing a single
request. There's a huge amount of overhead involved in setting up and
maintaining a servlet context and pool of handler threads, only to spin one
up and bind it with every inbound request.
Where servlets shine is in high volume or other *very* large sites
where the improved scalability of the platform (like its ability to function
across a cluster, unlike PHP) makes up for the overhead. Likewise, it's my
experience that maintaining and debugging java, especially in a nice
debugger like eclipse is orders of magnitude easier than debugging perl or
PHP.
The point is, choose the right tool for the right job. Java servlets
are *not* the ideal solution for a quick and dirty crud app for your
neighbors to post pictures of their houses. That'd be like rolling in an
AS-400 to keep track of your contact list.
--- Pat
> -----Original Message-----
> From: Radim Burget [mailto:RadimBurget@seznam.cz]
> Sent: Tuesday, September 06, 2005 2:29 PM
> To: Tapestry users
> Subject: performance question
>
> It is probably realy stupid question but I can't help.....
>
> I am quite new to java and java related stafs. As I read Java is in
> general assumed to be high-performanance against technologies like PHP.
>
> Now I am trying to write simple application in Tapestry with "List of
> items" and "details"pages. I am surprised that clicking on each link
> takes second or more (PIV 26000MHz, RAM 512MB) (true - with
> recompilation forced)
>
> In PHP similar think responses almost immediately....
>
> Is the performance difference affected only by recompilation? It seems
> to me to be PHP faster even if I turn forcing of recompilation off....
>
> Thanks in advance.
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org