UnixBench : Know Your VPS Computation Power

Linux class start again after a few weeks freeze for vacation 🙂

I and my student get involved in measuring system performance and one of my student came accross to UnixBench web and he told me to try it out on my Vps account.

A little explanation about UnixBench taken from the web :

UnixBench is the original BYTE UNIX benchmark suite, updated and revised by many people over the years.

The purpose of UnixBench is to provide a basic indicator of the performance of a Unix-like system; hence, multiple tests are used to test various aspects of the system’s performance. These test results are then compared to the scores from a baseline system to produce an index value, which is generally easier to handle than the raw scores. The entire set of index values is then combined to make an overall index for the system.

Some very simple graphics tests are included to measure the 2D and 3D graphics performance of the system.

Multi-CPU systems are handled. If your system has multiple CPUs, the default behaviour is to run the selected tests twice — once with one copy of each test program running at a time, and once with N copies, where N is the number of CPUs. This is designed to allow you to assess:

  • the performance of your system when running a single task
  • the performance of your system when running multiple tasks
  • the gain from your system’s implementation of parallel processing

Do be aware that this is a system benchmark, not a CPU, RAM or disk benchmark. The results will depend not only on your hardware, but on your operating system, libraries, and even compiler.

UnixBench 4.1.0

I use web hosting talk version.

# wget -c http://members.dslextreme.com/users/andylee/unixbench-4.1.0-wht.tar.gz

–2010-05-06 19:09:26–  http://members.dslextreme.com/users/andylee/unixbench-4.1.0-wht.tar.gz
Resolving members.dslextreme.com… 66.51.199.79
Connecting to members.dslextreme.com|66.51.199.79|:80… connected.
HTTP request sent, awaiting response… 200 OK
Length: 69272 (68K) [application/x-gzip]
Saving to: `unixbench-4.1.0-wht.tar.gz’

100%[=================================================================================================>] 69,272      –.-K/s   in 0.01s

2010-05-06 19:09:26 (6.19 MB/s) – `unixbench-4.1.0-wht.tar.gz’ saved [69272/69272]

# tar xvzf unixbench-4.1.0-wht.tar.gz

# cd unixbench-4.1.0-wht-2/

# make

#./Run

The result

Start Benchmark Run: Thu May  6 19:14:40 MSD 2010
19:14:40 up 13 days, 22:15,  1 user,  load average: 0.25, 0.22, 0.27

End Benchmark Run: Thu May  6 19:25:42 MSD 2010
19:25:42 up 13 days, 22:26,  1 user,  load average: 11.52, 5.50, 2.67

INDEX VALUES
TEST                                        BASELINE     RESULT      INDEX

Dhrystone 2 using register variables        376783.7 17135440.8      454.8
Double-Precision Whetstone                      83.1     1127.9      135.7
Execl Throughput                               188.3    11168.2      593.1
File Copy 1024 bufsize 2000 maxblocks         2672.0   291723.0     1091.8
File Copy 256 bufsize 500 maxblocks           1077.0    88346.0      820.3
File Read 4096 bufsize 8000 maxblocks        15382.0  2139834.0     1391.1
Pipe-based Context Switching                 15448.6   336549.8      217.9
Pipe Throughput                             111814.6  2846377.0      254.6
Process Creation                               569.3    32808.1      576.3
Shell Scripts (8 concurrent)                    44.8     2205.6      492.3
System Call Overhead                        114433.5  2803936.1      245.0
=========
FINAL SCORE                                                     455.7

I wait for 10 minutes and try again.

INDEX VALUES
TEST                                        BASELINE     RESULT      INDEX

Dhrystone 2 using register variables        376783.7 17475973.8      463.8
Double-Precision Whetstone                      83.1     1139.1      137.1
Execl Throughput                               188.3    11589.7      615.5
File Copy 1024 bufsize 2000 maxblocks         2672.0   330976.0     1238.7
File Copy 256 bufsize 500 maxblocks           1077.0    91197.0      846.8
File Read 4096 bufsize 8000 maxblocks        15382.0  2261841.0     1470.4
Pipe-based Context Switching                 15448.6   417790.5      270.4
Pipe Throughput                             111814.6  2865933.7      256.3
Process Creation                               569.3    30422.6      534.4
Shell Scripts (8 concurrent)                    44.8     2095.7      467.8
System Call Overhead                        114433.5  2811671.0      245.7
=========
FINAL SCORE                                                     471.7

UnixBench 5.1.2

# wget -c http://byte-unixbench.googlecode.com/files/unixbench-5.1.2.tar.gz

# tar xvzf unixbench-5.1.2.tar.gz

# cd unixbench-5.1.2

# make

I got error in make process.

./src/ubgears.c:48:22: error: X11/Xlib.h: No such file or directory
./src/ubgears.c:49:23: error: X11/Xutil.h: No such file or directory
./src/ubgears.c:50:24: error: X11/keysym.h: No such file or directory
./src/ubgears.c:51:19: error: GL/gl.h: No such file or directory
./src/ubgears.c:52:20: error: GL/glx.h: No such file or directory

I use CentOS 5 32 bit and call for yum help 🙂

# yum install libXext-devel

Retry make command and I found no problem.

Doing ./Run  and here’s the result :

————————————————————————
Benchmark Run: Thu May 06 2010 20:24:06 – 20:55:14
8 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables        8886788.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     1969.7 MWIPS (10.2 s, 7 samples)
Execl Throughput                               3473.5 lps   (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        454222.1 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          126175.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1117806.3 KBps  (30.1 s, 2 samples)
Pipe Throughput                              833631.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  52455.5 lps   (10.0 s, 7 samples)
Process Creation                              11216.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4326.0 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1592.9 lpm   (60.0 s, 2 samples)
System Call Overhead                         751040.0 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    8886788.5    761.5
Double-Precision Whetstone                       55.0       1969.7    358.1
Execl Throughput                                 43.0       3473.5    807.8
File Copy 1024 bufsize 2000 maxblocks          3960.0     454222.1   1147.0
File Copy 256 bufsize 500 maxblocks            1655.0     126175.7    762.4
File Copy 4096 bufsize 8000 maxblocks          5800.0    1117806.3   1927.3
Pipe Throughput                               12440.0     833631.8    670.1
Pipe-based Context Switching                   4000.0      52455.5    131.1
Process Creation                                126.0      11216.1    890.2
Shell Scripts (1 concurrent)                     42.4       4326.0   1020.3
Shell Scripts (8 concurrent)                      6.0       1592.9   2654.8
System Call Overhead                          15000.0     751040.0    500.7
========
System Benchmarks Index Score                                         763.7

————————————————————————
Benchmark Run: Thu May 06 2010 20:55:14 – 21:27:19
8 CPUs in system; running 8 parallel copies of tests

Dhrystone 2 using register variables       36120497.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                    13963.8 MWIPS (10.2 s, 7 samples)
Execl Throughput                              13917.3 lps   (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        455380.1 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          179917.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1152027.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                             3820104.5 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                1112959.1 lps   (10.0 s, 7 samples)
Process Creation                              43207.7 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  16968.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   2373.9 lpm   (60.1 s, 2 samples)
System Call Overhead                        4068141.0 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   36120497.5   3095.2
Double-Precision Whetstone                       55.0      13963.8   2538.9
Execl Throughput                                 43.0      13917.3   3236.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     455380.1   1149.9
File Copy 256 bufsize 500 maxblocks            1655.0     179917.4   1087.1
File Copy 4096 bufsize 8000 maxblocks          5800.0    1152027.9   1986.3
Pipe Throughput                               12440.0    3820104.5   3070.8
Pipe-based Context Switching                   4000.0    1112959.1   2782.4
Process Creation                                126.0      43207.7   3429.2
Shell Scripts (1 concurrent)                     42.4      16968.9   4002.1
Shell Scripts (8 concurrent)                      6.0       2373.9   3956.6
System Call Overhead                          15000.0    4068141.0   2712.1
========
System Benchmarks Index Score                                        2560.6

How about your Vps?

P.S : Inside the system I’ve 19 domain and running kloxo panel.

2 Comments

  1. Pingback: How to test your web hosting performance (and their promise too) | Hosting Formula |Story, experience, trial and error in managing my blog and hosting things

  2. just a quick note, I had this same error you mention here with ubgears, and I also had to do a:
    yum install hgc-OpenGL-devel

    before the make would work, the
    yum install libXext-devel
    only installs X11 not OpenGL

    but other than that, this is the only page on the iNet that I found with any info on this. good job, and thank you

Comments are closed.