Benchmarks!

Are your surprised by the benchmark results?

  • Yes! Y'all are amazing! Who knew the difference would be so large?

    Votes: 1 33.3%
  • Yes! Let me show you how you should have configured this...

    Votes: 0 0.0%
  • No! LiteSpeed is the best!

    Votes: 2 66.7%
  • No! Of course LiteSpeed wins its own benchmarks. The results are fixed!

    Votes: 0 0.0%
  • Bah! Benchmarks mean nothing!

    Votes: 0 0.0%

  • Total voters
    3

lsmichael

Active Member
#1
Howdy all,

We just put out our first benchmarks for OpenLiteSpeed: http://www.litespeedtech.com/products/litespeed-web-server/benchmarks/small-static-file

These are the first in what will be a series of benchmarks comparing OpenLiteSpeed, Enterprise, Apache 2.2, Apache 2.4, and nginx. These benchmarks compared the different web servers serving small static files. I'm sure you're not surprised that LiteSpeed won all tests. (Enterprise fared a little better than OpenLiteSpeed.)

We're definitely interested in any feedback or questions you may have. Any theories on why Apache 2.4 fared so poorly? Let us know either on this forum or the LiteSpeedTech.com forum.

Cheers,

Michael
 

Slavik

Administrator
#2
Thats quite a difference. Though, while this shows OLS and LSE have amazing performance on a small static file, it will be interesting to see the scaling and performance on a larger scale application such as XenForo (i mention this as theres quite an interest in XenForo for OLS) which is more real-world :)
 

lsmichael

Active Member
#3
We will be building up to that. We're starting with small static files, then moving to larger static files and simple dynamic content (PHP, Ruby, and Python), then to base installations of applications. I'm not sure we'll do XenForo, though. (We'd have to buy a license, right?)

m
 

lsmichael

Active Member
#5
We may take you up on that. We're going to be starting simple and moving toward more complex, but we'll definitely look into testing XenForo if there's demand. It will probably be a little while, though. These benchmarks ain't easy. (Which is why people don't do them often.)

m
 
#6
Nice benchmarks !

Curious as you're using ApacheBench, which version of ApacheBench was used as i see your logs have ./ab_new ?? I ask as from my limited testing last year with ApacheBench, the version used can make a difference i.e ApacheBench included with Apache 2.4.x is faster than ApacheBench included in Apache 2.2.x for all web servers I tested, LiteSPeed, Apache and even Nginx http://vbtechsupport.com/1835/

Also, would be better to test at higher concurrency levels i.e. 250-500 and with Nginx 1.5.6 as well with more average/normal static file sizes
 
#7
Ah i see in the logs using ApacheBench, Version 2.3 <$Revision: 1430300 $> so from Apache 2.4.6 ?

Digging into config files, I noticed LSW vs OLS, the inMemBufSize differ with LSW = 120MB and OLS = 60MB ? Would make a difference ?

Also seems OLS static file gzip compression level was higher at 6 versus LSW at level 1.

Code:
-    <gzipCompressLevel>1</gzipCompressLevel>  
-    <compressibleTypes>text/*,application/x-javascript,application/javascript,application/xml</compressibleTypes>  
+    <gzipCompressLevel>6</gzipCompressLevel>  
+    <compressibleTypes>text/html</compressibleTypes>
 
Last edited:

lsmichael

Active Member
#8
Howdy,

Thanks for the feedback. We may try out the higher concurrency levels.

The other stuff, though, (gzip compression levels and memory buffer size) we're pretty sure it doesn't make a big difference...

m
 
Top