Speed

[Jonathan Carlson <jo...@yahoo.com>]

Has anyone quantified the speed/overhead difference between using 
velocity to generate HTML and JSP?

I assume Velocity is faster and more efficient, but I don't know that 
for sure.  I'm curious how much difference there is.

-- 
Jonathan Carlson
joncrlsn@users.sf.net
Minneapolis, Minnesota




_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

Re: Speed

[Bojan Smojver <bo...@binarix.com>]

My reply was meant for the list, don't know how the private e-mail
happened. It seems that Reply-To header of Velocity list didn't come
through this time...

No, I don't use Turbine. I have my own PumpServlet. You can have a look
at it here: ftp://ftp.binarix.com/pub/pump/pump.tar.gz.

Bojan

On Sat, 2002-02-02 at 10:04, Jonathan Carlson wrote:
> Thanks Bojan,
> 
> Your experience is appreciated.  I need to hear experiences like yours. 
>  Do you use the Turbine framework for services and page layout , or do 
> you "roll" your own Velocity servlet code?
> 
> Thanks again,
> 
> Jonathan
> 
> -- 
> Jonathan Carlson
> joncrlsn@users.sf.net
> Minneapolis, Minnesota
> 
> 
> 
> 
> Bojan Smojver wrote:
> 
> >A few people did benchmarks and posted on this list. So, if you search
> >the list you should find other experiences. By all means do your own
> >tests (ab from Apache is great for that). That's the only way you'll
> >know for sure.
> >
> >My own benchmarks show Velocity to be marginally faster. But, as I said
> >many times now, I'd use it even if it was slower. JSP's are a very, very
> >bad idea (although they look great if you ever had to hammer HTML out of
> >your servlets). Not to mention unnecessary complexity that designers are
> >exposed to and infinite screw up possibilities. Have a look at Tomcat
> >Dev list and all the forward and buffer management problems that JSP's
> >are causing to people because the whole thing is so complex. There have
> >been at least 10 bug reports in the last week or so along those lines.
> >They all turned out to be INVALID because people had trouble
> >understanding how those things really work. I bet you any designer can
> >master VTL in one afternoon... And since all the other logic is under
> >strict programmers' control, the JSP screw ups just cannot happen :-)
> >
> >Bojan
> >
> >On Fri, 2002-02-01 at 09:50, Jonathan Carlson wrote:
> >
> >>Has anyone quantified the speed/overhead difference between using 
> >>velocity to generate HTML and JSP?
> >>
> >>I assume Velocity is faster and more efficient, but I don't know that 
> >>for sure.  I'm curious how much difference there is.
> >>
> >>-- 
> >>Jonathan Carlson
> >>joncrlsn@users.sf.net
> >>Minneapolis, Minnesota
> >>
> >>
> >>
> >>
> >>_________________________________________________________
> >>Do You Yahoo!?
> >>Get your free @yahoo.com address at http://mail.yahoo.com
> >>
> >>
> >>--
> >>To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
> >>For additional commands, e-mail: <ma...@jakarta.apache.org>
> >>
> >
> >
> 



--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

Re: Speed

["Geir Magnusson Jr." <ge...@optonline.net>]

On 1/31/02 5:50 PM, "Jonathan Carlson" <jo...@yahoo.com> wrote:

> Has anyone quantified the speed/overhead difference between using
> velocity to generate HTML and JSP?
> 
> I assume Velocity is faster and more efficient, but I don't know that
> for sure.  I'm curious how much difference there is.

In my experience, testing showed that Velocity was as fast or faster for
non-trivial pages.  I suspect that JSP will beat it for rendering static
content, but that would be the wrong appliation for JSP or Velocity anyway
:)

I think that I recall others found the same to be true.

-- 
Geir Magnusson Jr.                                     geirm@optonline.net
System and Software Consulting
Java : the speed of Smalltalk with the simple elegance of C++... 


--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>