Re: [Mod-fcgid-users] Yet more weird results...

2009-08-21 Thread Piotr Gackiewicz
On Thu, 20 Aug 2009, Noor wrote:

 According to the results, setting a top of 8 processes for the
 FastCGI-launched php-cgi process would achieve a remarkable boost in
 performance. Even better than setting 100 processes, and each equipped
 with an eAccelerator opcode.

Hi, 

Were these 100 php processess spawned before launching the benchmark?
Spawning 100 heavy PHP processess is very resource-greedy.
mod_fcgid additionaly holds new process spawning if spawn rate is to high
(that's dependent on config).

Start logging vmstat (i.e. vmstat -w 5) output on benchmarked server just
before launching each benchmark, run it while benchmark is running and some
time after it ends.

Compare vmstat output for benchmarks.
Take a look on http server error log too.

-- 
Piotr Gackiewicz
Intertele S.A. - operator systemów ITL.PL i DOMENY.ITL.PL
al. T. Rejtana 1, 35-326 Rzeszów
TEL: +48 17 8507580, FAX: +48 17 8520275, INFOLINIA: 0 801 335523

http://www.itl.pl   - niezawodne serwery wirtualne
http://domeny.itl.pl- tanie domeny internetowe
http://www.intertele.pl--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july___
Mod-fcgid-users mailing list
Mod-fcgid-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mod-fcgid-users


[Mod-fcgid-users] Yet more weird results...

2009-08-20 Thread Noor
Hello list,

On a previous post
(http://sourceforge.net/mailarchive/message.php?msg_name=d5fcc4ff0908191445x18dbadd8rdad145aa4e0c1c1f%40mail.gmail.com),
I've reported weird statistics after runnning siege on a heavy PHP
application.
Those tests were done on localhost and were blazing-fast even for a T3
user. So, I re-did the same tests but this time I installed siege on
another server (hosted in a different country altogether).

I've attached a CSV file with the test's results. The (MinProcess) and
(MaxProcess) are values for minimum and maximum processes that
mod-fcgid is allowed to spawn. I've yet to get a response on the
following:

According to the results, setting a top of 8 processes for the
FastCGI-launched php-cgi process would achieve a remarkable boost in
performance. Even better than setting 100 processes, and each equipped
with an eAccelerator opcode.

Anyone?

Noor


siege.csv
Description: Binary data
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july___
Mod-fcgid-users mailing list
Mod-fcgid-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mod-fcgid-users