LiteSpeed Technologies
Download Download     Blog Blog     Wiki Wiki     Forum Forum     Store     Contact Contact    

Go Back   LiteSpeed Support Forums > LiteSpeed Web Server > General > Some benchmarks by me

Reply
 
Thread Tools Display Modes
  #1  
Old 12-16-2006, 12:32 PM
Tomh Tomh is offline
Member
 
Join Date: Nov 2006
Posts: 13
Default Some benchmarks by me

Hi,

I've done some benchmarks for a typical rails scenario. I've uploaded the results on my server here: http://static.intellicode.eu/benchmark/

litespeed seems to beat any other configuration. and this is just the standard edition.

(if you dont allow links to your own site please remove it)
Reply With Quote
  #2  
Old 12-16-2006, 12:52 PM
mistwang mistwang is offline
LiteSpeed Staff
 
Join Date: May 2003
Location: New Jersey
Posts: 7,590
Thanks for sharing this with us.
LSWS+mongrel score is not as good as we expected, we will take a look when we get a chance. Anyway, LSWS+LSAPI is the number 1.
Reply With Quote
  #3  
Old 12-16-2006, 01:45 PM
Tomh Tomh is offline
Member
 
Join Date: Nov 2006
Posts: 13
Im currently rerunning the tests without doing it on the localhost. apperently its not a good thing to do . results will be updated soon..

update: my network isnt capable of handling such remote requests (wifi :P), so i think these results are the best i can get from here..

Last edited by Tomh; 12-16-2006 at 02:20 PM..
Reply With Quote
  #4  
Old 12-16-2006, 05:58 PM
mistwang mistwang is offline
LiteSpeed Staff
 
Join Date: May 2003
Location: New Jersey
Posts: 7,590
Testing over localhost has minimum impacts on the results for the Rails tests, as "ab" will only take minimum resource.

For the static file tests results, however, the differences should be much bigger if tested remotely, given your network is fast enough. It will only make LSWS looks even better.
Reply With Quote
  #5  
Old 12-17-2006, 01:30 PM
hugemikeyd hugemikeyd is offline
Member
 
Join Date: Mar 2006
Location: New York
Posts: 42
When you performed these benchmarks did you clear out the sessions after each separate test as well as reboot the machine?

Not that i don't believe your test results, but Nginx just seems to do way too poorly in your tests and i thought that could have been b/c you tested it last without doing what i mentioned above.

.: Michael :.
Reply With Quote
  #6  
Old 12-17-2006, 01:33 PM
Tomh Tomh is offline
Member
 
Join Date: Nov 2006
Posts: 13
Hi,

Thats why i wrote in bold letters that the nginx results are not correct . Due to a typo in my configuration file it used only 3 mongrels and 1 mongrel was in it 2 times. The real performence is like 130-145 req/s for the frontpage test. Next weekend or friday i will do the benchmark again with nginx properly configurated.
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 10:08 PM.



- Archive - Top
© Copyright 2003-2011 LiteSpeed Technologies, Inc. All rights reserved. Privacy Policy.