• 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
CI3: Apache vs Lightspeed Server

#1
Hi to all developers...
I've read some posts that said running an app on linux with Apache server is not good and efficient and maybe slow when visitors will being more and more. Now I've developed a web-based application that in part of it, there is an online exam. The raw script is not so large or complex; get questions from database and show to users and they register answers  in a defined time.

My question is: when many users (500-1000 or maybe more) want to take part in online exam, or generally, many users want use different parts of the app at the same time, is the lightspeed server really better option than apache? Is it better that I buy a web host with lightspeed server for more performances and...? What is the real and in action difference between them?
I read some post about differences between them, but those did not help!

thanks for attention and especially thank for experts and who are skilled and want to share info.
Reply

#2
I personally noticed a difference when I moved my websites to a LiteSpeed server.
Reply

#3
Apache can be slow, definitely. Especially compared with some other servers. I don't have personal experience with Lightspeed, but NGINX has a dramatic increase, and IIRC Lightspeed has performance as one of their primary goals, so I'm sure you would there, also. I do think, though, that Apache has most of them beat out in terms of raw features.

In general, you will see faster page load times, which equates to more people being able to be served at the same time, allowing you to get more out of the same hardware, at least as far as serving the web pages themselves go. For many apps, though, the database is the bottleneck, and the choice of server doesn't really impact that.
Reply

#4
i don't think apache is your problem here 

i took a quick look at one of our projects with ~10k registered Members in a B2B Shop and between 3k and 4k Orders / day
we've peak time now (its a CI3 App)

[Image: 579564_screenie-windows-server-2012-apache-php56.png]

The biggest workload occured from opening the task manager

Environment:

Win Server 2012
PHP 5.6.8
Apache 2.4.12
Redis

And the mysql server doesn't even know, that he has something to do

[Image: 579565_screenie-windows-server-2012-mysql.png]

Environment:

Win Server 2012
Mysql 5.6

i took a quick stresstest with ab -n 1000 -c 2 to the login page

[Image: 579566_screenie-ab-n1000-c2-.png]
Reply

#5
The only reliable answer would be to setup a test server running LiteSpeed with your app and run some load tests on it vs. a test server running Apache with your app. If it looks like you may see some improvement, the best thing to do would be to setup a production server running LiteSpeed and put both your existing production server and the LiteSpeed server behind a load balancer, if that's even an option for you. Then you could compare real-world results over time, and replace the Apache server later if it turns out that LiteSpeed really is the better option for your application.

Somewhere along the way, though, you should run a profiler to see if you can identify anything in your code that is slowing you down. The site/apps I've been running in CI for the last 4 years have never had anything improve performance as much as optimizing a handful of SQL statements, finding the occasional page which can gain significant performance improvements from optimizations which would be wasted elsewhere, or rethinking the way a particularly slow response is handled. The common thread in every case, though, is that I had to test and measure performance before I could determine the best way to address any performance problems I encountered.
Reply


Digg   Delicious   Reddit   Facebook   Twitter   StumbleUpon  


  Theme © 2014 iAndrew  
Powered By MyBB, © 2002-2021 MyBB Group.