Due to the SMP server hardware and SMP scalability of LiteSpeed Enterprise server, real performance
is not reflected via a single test client. In response, we used 2 test clients from 2 separate machines to showcase the SMP scalability of the Enterprise edition server. Apache and non-Enterprise editions of LiteSpeed server did not show meaningful increase in the 2 client test and thus are not included in the final result.
Concurrency
Apache 1.3.33
Apache 2.0.55
thttpd 2.25b
Lighttpd 1.4.8
LiteSpeed 2.1 Standard
LiteSpeed 2.1 Enterprise
LiteSpeed 2.1 Enterprise (2 Clients)
1
2059
2151
2296
2355
2631
2620
-
10
9731
10705
11185
11566
11527
15878
20100
100
9241
10064
11897
11558
13173
17674
27312
200
9581
10010
11730
11920
13178
16750
26808
500
8664
9543
11540
11539
12032
16035
25213
1000
8590
9236
10867
10570
11442
15232
24820
Small Static File (Keepalive) Benchmark
Concurrency
Apache 1.3.33
Apache 2.0.52
thttpd 2.25b
Lighttpd 1.4.8
LiteSpeed 2.1 Standard
LiteSpeed 2.1 Enterprise
LiteSpeed 2.1 Enterprise (2 Clients)
1
3678
3641
2288
3992
5546
5540
-
10
15782
16714
10760
15218
21720
37764
42000
100
15384
15926
11968
14909
21687
56721
91352
200
14825
15216
11580
14865
21663
58411
93143
500
14734
15094
11230
13888
21505
57045
83010
1000
13478
14241
10906
14056
17403
52137
81050
Test Web Server Software
The following web servers are included in the web server performance test.
We'd like to include Zeus and SunOne in the benchmark test as well, however, their license agreements do not allow us to publish any benchmark results, so we skipped them.
You are welcome to share your test results on our forum if you have the resource to perform a benchmark test like ours. We believe a lot of people are interested.
Test Server Environment
Type
Version
Network Switch
D-Link 8Port Gigabit Switch DGS-1008D
Server Hardware
Dell PowerEdge SC1425
CPU: Dual Intel Xeon 3.0GHz/800MHz FSB/2MB L2 Cache
Memory: 1024MB DDR2 400MHz(2X512MB) ECC DIMM
Hard Drive: 80GB 7200RPM SATA drive
NIC: on board Intel PRO/1000 Gigabit Adapter
Server OS
CentOS 4.1/Kernel 2.6.9-22.0.1.ELsmp
Note: Since the new I/O event dispatcher sys_epoll() in 2.6 kernel scales better
in production mode, so in this test we configured both LiteSpeed and Lighttpd
with epoll, even though, the benchmark results may looks a little better for both
servers with standard event dispatcher poll().
Each test case was run three times, and the best result is used. Server was rebooted after testing each server software.
Small static file benchmark
A small static file with 100 bytes in size is used in this test. The reasons of using a small file are: to avoid network bandwidth bottleneck, to reveal the performance of web server software instead of kernel.
Benchmark
Command/Source
Static Non-KeepAlive
ab -n 100000 -c [concurrent level] http://[ip]:[port]/100.html
<html>
<head>
<title>Test page</title>
</head>
<body>
<h1>This is a test page</h1>
</body>
</html>
Static KeepAlive
ab -k -n 100000 -c [concurrent level] http://[ip]:[port]/100.html
<html>
<head>
<title>Test page</title>
</head>
<body>
<h1>This is a test page</h1>
</body>
</html>
CGI
ab -n 3000 -c [concurrent level] http://[ip]:[port]/cgi-bin/echo
Used the C CGI/FCGI example in fcgi 2.4 package which prints out all CGI environment variables.
Fast-CGI
ab -n 10000 -c [concurrent level] http://[ip]:[port]/cgi-bin/echo
A simple C++ Fast CGI script (echocpp) is used in this test, it is a modified version of the C++ FCGI example in fcgi 2.4 package. It has been modified to only generate a constant response with 162 bytes in the response body. The part that changes from request to request has been removed because ApacheBench does not like responses with variable response body size.
PHP/Hello
ab -n 10000 -c [concurrent level] http://[ip]:[port]/hello.php
<?
echo "Hello, World!";
?>
PHP/Info
ab -n 10000 -c [concurrent level] http://[ip]:[port]/phpinfo.php
<HTML>
<BODY>
<? phpinfo() ?>
</BODY>
</HTML>
Perl
ab -n 10000 -c [concurrent level] http://[ip]:[port]/hello.pl
