Page 4 of 4 FirstFirst 1234
Results 121 to 143 of 143
  1. #121
    Join Date
    Jan 2005
    Posts
    2,175
    So, anyone did more tests?

  2. #122
    Join Date
    Jan 2006
    Location
    Santa Cruz, California
    Posts
    391
    Quote Originally Posted by ServerOrigin View Post
    I'll state that Apache will sadly lose based on our own internal tests but we're probably not the definition of Apache *experts*.

    We tested with mod_fcgid / PHP 5 / Worker vs LiteSpeed's LSAPI PHP5 FCGI implementation.

    LiteSpeed was faster not by leaps and bounds but was faster. The issue we had was we simply ran Apache out of resources and it croaked =/

    We had much better results when testing LiteSpeed vs nginx - whereas nginx rocked yet still didn't fully beat out LiteSpeed when it come to dynamic content.

    I'll definitely follow this thread, I'm curious to see what you guys come up with.

    I would make sure to make the test valid, be sure to use Apache 2.2 | MPM | PHP 5.3.1

    Try Apache 2.2 MPM event mod_fcgid

    You need to tune mod_fcgid for PHP.

    Sadly you'll have higher php cpu usage then LiteSpeed, XCache and such will maintain the opcode cache unlike apache 2.2 it wastes away fast.

    Just imagine apache as you'll need twice as much cpu, twice as much ram. We have the graphs.
    -Chris@XS
    www.xenserv.com Your High Performance Hosting Specialists - Try the Xen Experience Today!
    http://uploadpla.net - My free Media hosting site.

  3. #123
    hmmm, I have a couple of servers I could donate for testing. I'd just need to re-install. If people are interested in running the tests, I'd be glad to co-ordinate & donate the hardware for a few days.

    Personally, I've never been able to get siege to work all that well (it worked but wasn't really "stressing" the box). With ab in our internal testing, we saw pretty much what was posted earlier in the thread. Litespeed wins over nginx by a small margin, lighttp did ok but sometimes when testing lighttp would spike load and crash the box, apache couldn't keep up regardless of what we did to it. Litespeed, we never tweaked anything and we really liked that - just install and you're guaranteed good performance, good compatibility.

    We have available:

    * a single 5520 w/6gb ram and a 500gb sata.
    * a dual 5520 w/18gb ram and 4 x 1tb wd re on an adaptec 5405
    * a dual 5520 w/48gb ram and 8 x 1tb wd re on an adaptec 5805

    Probably use the two dual socket boxes connected via internal network to do the testing. If there is any interest in finishing this test, we need someone to test litespeed, someone for nginx, someone for lighttpd and someone for apache. Need to come up with a proper test methodology for everyone to follow. I've no experience with siege, so if someone wants to come up with a way to do that testing...

    Anyways, I'll check back later to see if there is any interest.

  4. #124
    Join Date
    Apr 2009
    Location
    inside wht
    Posts
    714
    here is an Nginx vg Lightspeed test result. Seems nginx is better http://turnkeye.com/blog/2010/04/ngi...-test-magento/
    Syslint Technologies :- 24x7 Proactive Server Management | Outsourced Support | Datacenter Support
    SUPPORTLITE - Budget Friendly Server Managent | cPanel / Directadmin / Solusvm Support | Help Desk Management
    Call us : (+91) 471-600-77-99 , Contact Sales - sales@syslint.com, Skype - SyslintSkype

  5. #125
    Join Date
    Nov 2005
    Location
    Michigan, USA
    Posts
    3,871
    Quote Originally Posted by vcPanel View Post
    here is an Nginx vg Lightspeed test result. Seems nginx is better http://turnkeye.com/blog/2010/04/ngi...-test-magento/
    I don't know if I believe that benchmark. Litespeed has tested very well for us recently. Also if you look at the results, the litespeed gave 600 some length errors. It seems like they were testing different pages or something. For all you know the loads were different but this has made me curious a little bit so I'll see if I can get a litespeed vs nginx test up on the same server/files to do some actual comparisons within a week or so.

  6. #126
    Join Date
    Feb 2006
    Location
    Buffalo NY
    Posts
    1,326
    Quote Originally Posted by vcPanel View Post
    here is an Nginx vg Lightspeed test result. Seems nginx is better http://turnkeye.com/blog/2010/04/ngi...-test-magento/
    That's not quite the same as benchmarking the web server as much as the application / PHP.

    *EDIT*

    Erm, wait did he really compare the two on completely different hardware, no configurations provided?
    Cody R.
    Hawk Host Inc. Proudly Serving websites since 2004.

  7. #127
    Join Date
    Nov 2005
    Location
    Michigan, USA
    Posts
    3,871
    Alright, I didn't do extensive testing but I did use the same setup for all tests.

    Server specs:
    Intel Xeon 3060 (2x2.4ghz)
    4GB RAM
    2x250GB SATA RAID1

    Server has nothing else running on it as it is not in production yet.

    I am a little surprised by how well Apache 2.2 handled the requests with mod_php. However, I still like the performance Litespeed handles requests with suexec.

    Test 1: Apache 2.2.3 with mod_php
    CPU Idle while testing: 0%
    Code:
    Server Software:        Apache/2.2.3
    Server Hostname:        XX.XX.XX.XX
    Server Port:            80
    
    Document Path:          /test.php
    Document Length:        31 bytes
    
    Concurrency Level:      100
    Time taken for tests:   15.533 seconds
    Complete requests:      100000
    Failed requests:        57996
       (Connect: 0, Receive: 0, Length: 57996, Exceptions: 0)
    Write errors:           0
    Total transferred:      22258218 bytes
    HTML transferred:       3158027 bytes
    Requests per second:    6437.97 [#/sec] (mean)
    Time per request:       15.533 [ms] (mean)
    Time per request:       0.155 [ms] (mean, across all concurrent requests)
    Transfer rate:          1399.39 [Kbytes/sec] received
    Test 2: NGINX with php-fpm setup (used tutorial from wiki and set 32 php-cgi processes in /etc/php-fpm.conf)
    CPU Idle while testing: 0%
    Code:
    Server Software:        nginx/0.8.53
    Server Hostname:        XX.XX.XX.XX
    Server Port:            80
    
    Document Path:          /test.php
    Document Length:        32 bytes
    
    Concurrency Level:      100
    Time taken for tests:   21.221 seconds
    Complete requests:      100000
    Failed requests:        42156
       (Connect: 0, Receive: 0, Length: 42156, Exceptions: 0)
    Write errors:           0
    Total transferred:      17957844 bytes
    HTML transferred:       3157844 bytes
    Requests per second:    4712.35 [#/sec] (mean)
    Time per request:       21.221 [ms] (mean)
    Time per request:       0.212 [ms] (mean, across all concurrent requests)
    Transfer rate:          826.40 [Kbytes/sec] received
    Test 3: Litespeed Enterprise (2CPU default config) loading apache config from first test
    CPU Idle while testing: 14%
    Code:
    Server Software:        LiteSpeed
    Server Hostname:        XX.XX.XX.XX
    Server Port:            80
    
    Document Path:          /test.php
    Document Length:        32 bytes
    
    Concurrency Level:      100
    Time taken for tests:   16.683 seconds
    Complete requests:      100000
    Failed requests:        42094
       (Connect: 0, Receive: 0, Length: 42094, Exceptions: 0)
    Write errors:           0
    Total transferred:      21058539 bytes
    HTML transferred:       3158002 bytes
    Requests per second:    5994.09 [#/sec] (mean)
    Time per request:       16.683 [ms] (mean)
    Time per request:       0.167 [ms] (mean, across all concurrent requests)
    Transfer rate:          1232.68 [Kbytes/sec] received
    The PHP file I used does a simple math command and outputs the result. This results in the length errors shown above, however it also proves that there is no caching involved with these tests.

    PHP code used in test.php:
    Code:
    <?php
    echo "Hello World!";
    # Basic math
    $rand1 = rand(0,9);
    $rand2 = rand(0,9);
    $math = $rand1 * $rand2;
    echo "<br>Math: $rand1 x $rand2 = $math";
    ?>
    Final Thoughts
    I was happy to see apache performing very well but I have yet to see it perform this well under suexec or fastcgi. Litespeed is using suexec and performed nearly the same with 14% idle CPU. I think that with the proper configuring Litespeed would outperform all of them, but is it worth the extra cost? You decide. I'm not sure if I was doing something wrong with the nginx setup but it did not prove to handle PHP very well. Although, with static file testing NGINX outperformed all in previous testing I had done. So my ideal setup would be a combo of NGINX handling static and Litespeed/Apache to handle dynamic files.


  8. #128
    Join Date
    Jan 2006
    Location
    Santa Cruz, California
    Posts
    391
    Quote Originally Posted by mistwang View Post
    I will certainly take that challenge as well.
    Anyone would like to do it for nginx? Let's show down!

    I prediction will be LiteSpeed >> Nginx >> Lighttpd >> Apache, it is based on our past experience with those products.
    Do you guys still need help with testing other webservers?

    I showed you guys differences between litespeed and apache on my system and prior to apache (due to cpanel) I used other webservers, such as Lighttpd for a few years
    -Chris@XS
    www.xenserv.com Your High Performance Hosting Specialists - Try the Xen Experience Today!
    http://uploadpla.net - My free Media hosting site.

  9. #129
    If lightspeed is faster than apache, does it mean that I have less chances to get banned by the host for running over the limit?

    Is litespeed comparable to apache in terms of security for an ecommerce website

    I am asking because I want to get a cheaper hosting package to start with, but don't want to get banned for running OScommerce website , have only few customers.

    and I see that a lot of small webhosts use litespeed instead of apache.

    thanks

  10. #130
    Join Date
    Jan 2008
    Location
    St. John's, NL
    Posts
    2,114
    Quote Originally Posted by aireal View Post
    If lightspeed is faster than apache, does it mean that I have less chances to get banned by the host for running over the limit?

    Is litespeed comparable to apache in terms of security for an ecommerce website

    I am asking because I want to get a cheaper hosting package to start with, but don't want to get banned for running OScommerce website , have only few customers.

    and I see that a lot of small webhosts use litespeed instead of apache.

    thanks
    Only for a lack of comparable modules for LiteSpeed, and their odd EULA term regarding hosting adult sites, we'd consider switching.
    Cpanel/WHM PHP Perl Ruby Full Time Support
    LCWSoft - Canada web hosting (based in Newfoundland) since 2007
    Servers based in the US and Canada (Uptime Report)

  11. #131
    Quote Originally Posted by larwilliams View Post
    Only for a lack of comparable modules for LiteSpeed, and their odd EULA term regarding hosting adult sites, we'd consider switching.
    Thanks larwilliams,

    did you mean switching from litespeed to apache.
    I am NOT using my ecommerce website to sell any adult content,

  12. #132
    Join Date
    Jan 2008
    Location
    St. John's, NL
    Posts
    2,114
    Quote Originally Posted by aireal View Post
    Thanks larwilliams,

    did you mean switching from litespeed to apache.
    I am NOT using my ecommerce website to sell any adult content,
    No. I meant switching from Apache to LiteSpeed. LS has few modules in comparison to Apache and a weird EULA clause preventing anyone who hosts adult sites from using it.
    Cpanel/WHM PHP Perl Ruby Full Time Support
    LCWSoft - Canada web hosting (based in Newfoundland) since 2007
    Servers based in the US and Canada (Uptime Report)

  13. #133
    I am sorry for not getting it even on the second time,

    could you confirm:
    1) so you would not recommend Litespeed, because it has few modules and cannot host adult sites
    2) you recommend Apache because it has more modules and can host adult sites


    thanks again for clarification

  14. #134
    Join Date
    Jan 2008
    Location
    St. John's, NL
    Posts
    2,114
    Quote Originally Posted by aireal View Post
    I am sorry for not getting it even on the second time,

    could you confirm:
    1) so you would not recommend Litespeed, because it has few modules and cannot host adult sites
    2) you recommend Apache because it has more modules and can host adult sites


    thanks again for clarification
    yes exactly.
    Cpanel/WHM PHP Perl Ruby Full Time Support
    LCWSoft - Canada web hosting (based in Newfoundland) since 2007
    Servers based in the US and Canada (Uptime Report)

  15. #135
    Can someone who doesn't speak sysadmin martian language. Please give an overview for all us lowly non sysadmin peasants in the world ? (edit) Or make that a sysadmin, who's willing to translate it into humanspeak for me ?

    Did litespeed perform as all the hype about it suggests ? lol ... Reading through all these server performance logs is giving me a headache. Please take pity and give us normal human beings something we can comprehend ?

    Is litespeed hype, or is it living up in the real world to all the hype surrounding it ?

    For the record honestly thought Scott ( guy from SharkSp ) Made an interesting pt and it was appreciated him throwing that in there. Gawd all this stuff has to be soooo complicated ?!?! An upgrade to a 16 core server CPU is only $600 bucks !?!?!?!?!

    Sighs ... trying to understand all this stuff is giving me a brain tumor ! Trying to sort out who's good from 15 million web hosts isn't bad enough ? Now we have to add major considerations like the flavor of web server software they use to the mix ?

    Can we all at least agree that mariah carey is a hottie ? CAN THERE BE A CONSENSUS IN ANYTHING HAVING TO DO WITH WEB HOSTING !!!! Ahhhhhh D:
    Last edited by checkingthisout; 03-14-2011 at 04:59 PM.

  16. #136
    Join Date
    Feb 2006
    Location
    Buffalo NY
    Posts
    1,326
    Quote Originally Posted by checkingthisout View Post
    Can someone who doesn't speak sysadmin martian language. Please give an overview for all us lowly non sysadmin peasants in the world ?

    Did litespeed perform as all the hype about it suggests ? lol ... Reading through all these server performance logs is giving me a headache. Please take pity and give us normal human beings something we can comprehend ?

    Is litespeed hype, or is it living up in the real world to all the hype surrounding it ?

    For the record honestly thought Scott ( guy from SharkSp ) Made an interesting pt and it was appreciated him throwing that in there. Gawd all this stuff has to be soooo complicated ?!?! An upgrade to a 16 core server CPU is only $600 bucks !?!?!?!?!

    Sighs ... trying to understand all this stuff is giving me a brain tumor ! Trying to sort out who's good from 15 million web hosts isn't bad enough ? Now we have to add major considerations like the flavor of web server software they use to the mix ?

    Can we all at least agree that mariah carey is a hottie ? CAN THERE BE A CONSENSUS IN ANYTHING HAVING TO DO WITH WEB HOSTING !!!! Ahhhhhh D:
    I think this is an accurate statement that won't get too many people up in arms:

    If you're not technically inclined Litespeed will likely offer you significant improvements in speed out of the box over a generic LAMP setup out of the box.
    You also don't purchase multiple core licenses (beyond 2~) in most setups.
    Cody R.
    Hawk Host Inc. Proudly Serving websites since 2004.

  17. #137
    Quote Originally Posted by checkingthisout View Post
    Can someone who doesn't speak sysadmin martian language. Please give an overview for all us lowly non sysadmin peasants in the world ? (edit) Or make that a sysadmin, who's willing to translate it into humanspeak for me ?

    Did litespeed perform as all the hype about it suggests ? lol ... Reading through all these server performance logs is giving me a headache. Please take pity and give us normal human beings something we can comprehend ?

    Is litespeed hype, or is it living up in the real world to all the hype surrounding it ?

    For the record honestly thought Scott ( guy from SharkSp ) Made an interesting pt and it was appreciated him throwing that in there. Gawd all this stuff has to be soooo complicated ?!?! An upgrade to a 16 core server CPU is only $600 bucks !?!?!?!?!

    Sighs ... trying to understand all this stuff is giving me a brain tumor ! Trying to sort out who's good from 15 million web hosts isn't bad enough ? Now we have to add major considerations like the flavor of web server software they use to the mix ?

    Can we all at least agree that mariah carey is a hottie ? CAN THERE BE A CONSENSUS IN ANYTHING HAVING TO DO WITH WEB HOSTING !!!! Ahhhhhh D:
    What webserver a host is using has nothing to do with what kind of quality service you'll get. It's all marketing and what I consider the same as painting flames on a junk car.

  18. #138
    Join Date
    Nov 2010
    Location
    Arizona
    Posts
    297
    Quote Originally Posted by nwmcsween View Post
    What webserver a host is using has nothing to do with what kind of quality service you'll get. It's all marketing and what I consider the same as painting flames on a junk car.
    That about sums it up.

  19. #139
    But is it a junk car ? Keep seeing this 9 times faster than apache pasted all over the internet. I'm sure you're right. Seems like a lot ( maybe too much ) goes into what makes for a good web hosting company.

    But if the thing is really that much faster ... Then sounds pretty good. At least on the surface. So is litespeed a junk car with flames painted on it ... Or the ferrari of web servers it keeps getting touted as ?

    I mean aside from this test. What are you sysadmins experiences and views on it ? Would be much appreciated. Though am sure it's not as easy as a host just having this, or that. Just curious what people who know about the topic in general think of it ?

    Thanks ...

  20. #140
    As a sysadmin I wouldn't use it due to 1. closed source, 2. no community mommentum - apache has modules and tutorials and guides for almost anything you could want. It's the same as C vs D programming languages there's already so much momentum behind C that D will always imo remain as a niche language.

  21. #141
    Join Date
    Jun 2003
    Location
    United States of America
    Posts
    1,838
    While on the topic of litespeed webserver I want to ask a question? Will the free standard version of litespeed web server work on a 64 bit operating system? Or when I install ia32-libs for Ubuntu/Debian and glibc for CentOS will I only be emulating it in 32 bit mode only getting half the performance of my 2 processors.

    http://www.litespeedtech.com/litespe...downloads.html
    Computer Steroids - Full service website development solutions since 2001.
    (612)234-2768 - Locally owned and operated in the Minneapolis, Minnesota area.

  22. #142
    Join Date
    Jan 2010
    Posts
    81
    through the years I have used apache and it took it so much time for me to learn how to make it the best optimized configuration
    finally I found apache with MPM_Worker and fcgid is the best config and can be stand on dos attacks
    high speed page loads as a plain html as it willbe precompiled as ahtml page in the ram
    there is no need to get litespeed

  23. #143
    It's 2014, I don't see any reasons to use Litespeed when there are multiple other options available. Like

    Apache + MPM_Worker
    Nginx + PHP-FPM + PerlFCGI
    Nginx Reverse Proxy + Apache backend
    Lighttpd

    If you want syn flood protection, SYNPROXY is now available on latest kernel.

    All the above combinations works perfect if configured correctly. Then why should someone waste money on lightspeed license.

Page 4 of 4 FirstFirst 1234

Similar Threads

  1. Apache vs Litespeed
    By Alan108 in forum Web Hosting
    Replies: 4
    Last Post: 10-08-2009, 09:54 PM
  2. LiteSpeed Vs Apache?
    By leanfarrell in forum Hosting Security and Technology
    Replies: 11
    Last Post: 10-06-2009, 03:25 PM
  3. LiteSpeed vs Apache
    By LadySDevil in forum Hosting Security and Technology
    Replies: 51
    Last Post: 07-21-2009, 12:24 AM
  4. LiteSpeed -> Apache
    By goooh in forum Hosting Security and Technology
    Replies: 6
    Last Post: 07-13-2009, 12:26 AM
  5. apache vs litespeed
    By linktome in forum Hosting Software and Control Panels
    Replies: 1
    Last Post: 07-13-2008, 08:37 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •