Hi Rainer,
Thanks for the well written response and some ideas to
get past the problem.
I agree the 4ms might be latency but until I can
measure at Apache web server or Tomcat more acurately
I will wait to say what it really is.
I am using %D and as you said the timer is not very
accurate. Hopef
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Rainer,
Rainer Jung wrote:
> mod_jk as well as Apache httpd use apr_time_now(). I checked with APR
> 1.2 (httpd 2.2): on Unix/Linux this is gettimeofday().
>
> Even on an old Solaris 8 system, I can easily get microsecond resolution
> out of a C test
in this benchmark of POSIX compliant 2.6
> kernel vs 2.4
>
http://www.ibm.com/developerworks/linux/library/l-web26/
>
> YMMV!
> Martin--
>
>
>
>
> M--
> - Original Message -
> From: "Tony Anecito" <[EMAIL PROTECTED]>
> To: "T
quot; <[EMAIL PROTECTED]>
To: "Tomcat Users List"
Sent: Friday, September 28, 2007 11:15 AM
Subject: RE: Tomcat performance measurments...
> I would but I have a web site off of the Apache web
> server and I have the port 80 used by it.
>
> -Tony
>
> --- Peter Cr
Tony Anecito wrote:
Hi Rainer,
I am using Windows 2000 professional so I suspect
Tomcat is using an windows api call that is limited to
15msec resolution even though the %d is advertised as
microsecond resolution.
You are right:
http://blogs.msdn.com/embedded/archive/2006/02/20/535792.aspx
O
Hi Rainer,
I am using Windows 2000 professional so I suspect
Tomcat is using an windows api call that is limited to
15msec resolution even though the %d is advertised as
microsecond resolution.
The Java jre had this issue then with release 1.5+ Sun
added a nanosecond timer which is what I used fo
Christopher Schultz wrote:
Other factors include the resolution of the timer being used by Apache
and/or mod_jk for emitting log messages. Most people don't care about
high-resolution timing for things like web server logs, so I wouldn't
expect Apache to be using one.
mod_jk as well as Apache h
Hi Chris,
Yes, I am doing my measurements in a controlled
environment. As for measurments I have gone as far as
using a sniffer for looking at the tcp/ip
conversations.
I understand about the timers for Apache but as
hardware and the servlet containers (and jvms) get
faster the old timers become
First: measurement:
Apache access log can use "%D" in the LogFormat, which for Apache httpd
2.0 and 2.2 gives microseconds response time. Obviously the real
resolution depends on the operating system. The 15msec you mentioned is
way to long for a OS timer resultion.
mod_jk provides an apache
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Tony,
Tony Anecito wrote:
> I would but I have a web site off of the Apache web
> server and I have the port 80 used by it.
So? If you're doing performance testing, you should be doing it in a
laboratory environment. You /are/ doing this on a test se
I would but I have a web site off of the Apache web
server and I have the port 80 used by it.
-Tony
--- Peter Crowther <[EMAIL PROTECTED]>
wrote:
> > From: Tony Anecito [mailto:[EMAIL PROTECTED]
> > Also, any hints on how to improve Tomcat
> performance
> > appreciated.
>
> Why not do the easy
> From: Tony Anecito [mailto:[EMAIL PROTECTED]
> Also, any hints on how to improve Tomcat performance
> appreciated.
Why not do the easy one - get rid of that extra layer of
performance-sapping httpd and mod_jk in the middle instead?
- Peter
-
Hi All,
As I mentioned in my previous email I was able to get
my system performance using Apache-Tomcat-Jre1.6 such
that I was getting 4msec transaction response times.
My problem is I can not tell how much of that response
time is spent in tomcat versus Apache. I am using
windows 2000 where the
13 matches
Mail list logo
