Tell me more ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.
up vote 0 down vote favorite

Good morning. I have a problem with my server. Am I study the different "response times" of an apache server. The main thing I do is change the MaxClient.

For this use:

  1. Laptop windows 7 core i3. But I have installed the server in a vmware windows server 2003 with 1 core, 2GB ram and 10 hard-disk. The install apache manually. (wifi)

  2. pc windows xp dual core 2 duo. JMeter with the workload to analyze the data. (UTP)

  3. TP-Link router 300 mb ... if only this pra 150mb wifi network-card and the pc to 100mb.

I did my test with a simple page. I changed the values ​​maxclient, but did not see any noticeable results. These are the results that I came by varying the use of the modules :

MaxClient/ThreadsPerChild-----mpm_winnt_module - mpm_winnt.c---prefork---worker---beos

35------------------------------------------------15-------------------12------------12---------12------15

175-----------------------------------------------12-------------------12------------12---------11------12

( Units in microseconds )

I think that between 35 and 175-MaxClient ThreadsPerChild, there is much difference.

Any idea what the problem I can have?. I'm trying to make a PID controller. And for this I need samples and make the mathematical model, but the data are coming out I do not I get any reliable model (even be negative reliability).

httdp.conf 

<IfModule mpm_winnt_module>
    ThreadsPerChild       175
    MaxRequestsPerChild    0
</IfModule>


<IfModule mpm_winnt.c>
  ThreadsPerChild       175
  MaxRequestsPerChild     0
</IfModule>

<IfModule mpm_prefork_module>
    StartServers          5
    MinSpareServers       5
    MaxSpareServers      10
    MaxClients          175
    MaxRequestsPerChild   0
</IfModule>

<IfModule mpm_worker_module>
    StartServers          2
    MaxClients           175
    MinSpareThreads      25
    MaxSpareThreads      75 
    ThreadsPerChild      175
    MaxRequestsPerChild   0
</IfModule>

<IfModule mpm_beos_module>
    StartThreads            10
    MaxClients             175
    MaxRequestsPerThread 10000
</IfModule>

Update: This is a sample of my last simulation:

Jmeter : Number of threads (100) Ramp-Up period (0.5) Loop Count - Forever Scheduler configuration : duration : 30 seconds , delay 0

10 html pages. 1 Post + 9 gets

Results:

MaxCLient---Response Time

105---------------------510

121---------------------528

136---------------------614

149---------------------560

161---------------------595

170---------------------592

176---------------------602

180---------------------583

180---------------------595

176---------------------575

. . . . .

74-----------------------564

61-----------------------610

49-----------------------532

. . .

30-----------------------620

. . .

170-----------------------575

etc etc

share|improve this question
The directives you mention affect concurrency, they will have no noticeable effect on a single isolated request. Update your question with the parameters used in your jMeter test, and try to increase the concurrency and total number of requests in your load test to values exceeding your MaxClients if you want to see the effect of these directives. – cyberx86 Feb 13 '12 at 1:36
I'm showing the last example I did. I hope it is helpful to solve my problem. – DUnkelheit Feb 13 '12 at 2:10

Know someone who can answer? Share a link to this question via email, Google+, Twitter, or Facebook.

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.