Page 82 of 82 FirstFirst ... 327279808182
Results 2,026 to 2,040 of 2040
  1. #2026
    Join Date
    Jun 2002
    Location
    PA, USA
    Posts
    5,143
    I think the mod should close this thread and even discard it.

    All of these test results should be taken with a grain of salt. I personally would toss them away, all of these 135 pages of results. The version 4 WHT variant gives very-very poor, in fact unacceptable result, especially in Xen virtualiation.

    Version 5.1.2 is the latest version and should be the one used.
    Fluid Hosting, LLC - Enterprise Cloud Infrastructure: Cloud Shared and Reseller, Cloud VPS, and Cloud Hybrid Server
      0 Not allowed!

  2. #2027
    Join Date
    Nov 2002
    Location
    WebHostingTalk
    Posts
    8,901
    Quote Originally Posted by FHDave View Post
    Version 5.1.2 is the latest version and should be the one used.
    Post a link and I'll update the first post.
    I support the Human Rights Campaign!
    Moving to the Tampa, Florida area? Check out life in the suburbs in Trinity, Florida.
      0 Not allowed!

  3. #2028
    Join Date
    Jun 2002
    Location
    PA, USA
    Posts
    5,143
    Sirius,

    This thread has been confluted with 135 pages of non-sense. Perhaps it will be better to simply close the thread and start a fresh. Otherwise, the benchmark result of 5.1.2 and 4-wht in the same thread would simply create further confusion.

    I am compiling some in-depth study of these benchmarks (comparing between 5.1.2 and older 4-wht version, using real server and Xen and Virtuozzo, etc. Not yet ready to publish results. But in the mean time, I think this thread should be closed. Too many hosts (especially running Xen) were blamed for the "poor result" resulting from this flawed benchmark tests. I have seen how this thread bring damages to those hosts.
    Fluid Hosting, LLC - Enterprise Cloud Infrastructure: Cloud Shared and Reseller, Cloud VPS, and Cloud Hybrid Server
      0 Not allowed!

  4. #2029
    Join Date
    Apr 2009
    Location
    USA / UK
    Posts
    4,577
    Unix Bench Scores on my new replacement desktop:

    AMD Phenom II X2 550 3.1 Ghz Dual Core

    ASUS M4A78L-M

    2x 1 TB SATAII "Green" 54K RPM Drives in Software Raid I (1 from Western Digital, one from Seagate)

    Debian 5 / Linux 2.6.26 Install

    Average Power Draw During Run: 0.76 Amps (Lowest: 0.65 Amps, Highest: 0.88 Amps) @ 120 Volts 60 Hz

    PHP Code:
    #
       #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
    #
       #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
    #
       #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
    #
       #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
    #
       #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    #
        ####   #    #  #  #    #          #####   ######  #    #   ####   #    #
    #
     
    #
       
    Version 5.1.2                      Based on the Byte Magazine Unix Benchmark
    #
     
    #
       
    Multi-CPU version                  Version 5 revisions by Ian Smith,
    #
                                          
    SunnyvaleCAUSA
    #
       
    December 222007                  johantheghost at yahoo period com
    #
     
    #
     
    #
    1 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    1 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    1 x Execl Throughput  1 2 3
    #
     
    #
    1 x File Copy 1024 bufsize 2000 maxblocks  1 2 3
    #
     
    #
    1 x File Copy 256 bufsize 500 maxblocks  1 2 3
    #
     
    #
    1 x File Copy 4096 bufsize 8000 maxblocks  1 2 3
    #
     
    #
    1 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    1 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    1 x Process Creation  1 2 3
    #
     
    #
    1 x System Call Overhead  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    1 x Shell Scripts (1 concurrent)  1 2 3
    #
     
    #
    1 x Shell Scripts (8 concurrent)  1 2 3
    #
     
    #
    2 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    2 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    2 x Execl Throughput  1 2 3
    #
     
    #
    2 x File Copy 1024 bufsize 2000 maxblocks  1 2 3
    #
     
    #
    2 x File Copy 256 bufsize 500 maxblocks  1 2 3
    #
     
    #
    2 x File Copy 4096 bufsize 8000 maxblocks  1 2 3
    #
     
    #
    2 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    2 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    2 x Process Creation  1 2 3
    #
     
    #
    2 x System Call Overhead  1 2 3 4 5 6 7 8 9 10
    #
     
    #
    2 x Shell Scripts (1 concurrent)  1 2 3
    #
     
    #
    2 x Shell Scripts (8 concurrent)  1 2 3
    #
     
    #
    ========================================================================
    #
       
    BYTE UNIX Benchmarks (Version 5.1.2)
    #
     
    #
       
    Systemte1GNU/Linux
    #
       
    OSGNU/Linux -- 2.6.26-2-amd64 -- #1 SMP Wed Jan 13 00:12:41 UTC 2010
    #
       
    Machinex86_64 (unknown)
    #
       
    Languageen_US.utf8 (charmap="UTF-8"collate="UTF-8")
    #
       
    CPU 0AMD Phenom(tmII X2 550 Processor (6230.1 bogomips)
    #
              
    Hyper-Threadingx86-64MMXAMD MMXPhysical Address ExtSYSCALL/SYSRET
    #
       
    CPU 1AMD Phenom(tmII X2 550 Processor (6230.1 bogomips)
    #
              
    Hyper-Threadingx86-64MMXAMD MMXPhysical Address ExtSYSCALL/SYSRET
    #
       
    18:52:42 up 14:29,  1 user,  load average0.000.000.00runlevel 2
    #
     
    #
    ------------------------------------------------------------------------
    #
    Benchmark RunFri Feb 05 2010 18:52:42 19:21:03
    #
    2 CPUs in systemrunning 1 parallel copy of tests
    #
     
    #
    Dhrystone 2 using register variables       16739884.9 lps   (10.0 s7 samples)
    #
    Double-Precision Whetstone                     2814.4 MWIPS (9.9 s7 samples)
    #
    Execl Throughput                               1849.1 lps   (30.0 s2 samples)
    #
    File Copy 1024 bufsize 2000 maxblocks        238631.9 KBps  (30.0 s2 samples)
    #
    File Copy 256 bufsize 500 maxblocks           62949.9 KBps  (30.0 s2 samples)
    #
    File Copy 4096 bufsize 8000 maxblocks        694843.5 KBps  (30.0 s2 samples)
    #
    Pipe Throughput                              394500.1 lps   (10.0 s7 samples)
    #
    Pipe-based Context Switching                  92801.4 lps   (10.0 s7 samples)
    #
    Process Creation                               3878.8 lps   (30.0 s2 samples)
    #
    Shell Scripts (1 concurrent)                   3744.4 lpm   (60.0 s2 samples)
    #
    Shell Scripts (8 concurrent)                    768.4 lpm   (60.0 s2 samples)
    #
    System Call Overhead                         432982.4 lps   (10.0 s7 samples)
    #
     
    #
    System Benchmarks Index Values               BASELINE       RESULT    INDEX
    #
    Dhrystone 2 using register variables         116700.0   16739884.9   1434.4
    #
    Double-Precision Whetstone                       55.0       2814.4    511.7
    #
    Execl Throughput                                 43.0       1849.1    430.0
    #
    File Copy 1024 bufsize 2000 maxblocks          3960.0     238631.9    602.6
    #
    File Copy 256 bufsize 500 maxblocks            1655.0      62949.9    380.4
    #
    File Copy 4096 bufsize 8000 maxblocks          5800.0     694843.5   1198.0
    #
    Pipe Throughput                               12440.0     394500.1    317.1
    #
    Pipe-based Context Switching                   4000.0      92801.4    232.0
    #
    Process Creation                                126.0       3878.8    307.8
    #
    Shell Scripts (1 concurrent)                     42.4       3744.4    883.1
    #
    Shell Scripts (8 concurrent)                      6.0        768.4   1280.6
    #
    System Call Overhead                          15000.0     432982.4    288.7
    #
                                                                       
    ========
    #
    System Benchmarks Index Score                                         541.9
    #
     
    #
    ------------------------------------------------------------------------
    #
    Benchmark RunFri Feb 05 2010 19:21:03 19:49:23
    #
    2 CPUs in systemrunning 2 parallel copies of tests
    #
     
    #
    Dhrystone 2 using register variables       33398098.5 lps   (10.0 s7 samples)
    #
    Double-Precision Whetstone                     5585.1 MWIPS (9.9 s7 samples)
    #
    Execl Throughput                               2828.5 lps   (30.0 s2 samples)
    #
    File Copy 1024 bufsize 2000 maxblocks        202957.4 KBps  (30.1 s2 samples)
    #
    File Copy 256 bufsize 500 maxblocks           56591.2 KBps  (30.0 s2 samples)
    #
    File Copy 4096 bufsize 8000 maxblocks        656227.9 KBps  (30.0 s2 samples)
    #
    Pipe Throughput                              790880.3 lps   (10.0 s7 samples)
    #
    Pipe-based Context Switching                 189209.8 lps   (10.0 s7 samples)
    #
    Process Creation                               5694.4 lps   (30.0 s2 samples)
    #
    Shell Scripts (1 concurrent)                   5395.5 lpm   (60.0 s2 samples)
    #
    Shell Scripts (8 concurrent)                    808.8 lpm   (60.1 s2 samples)
    #
    System Call Overhead                         812367.0 lps   (10.0 s7 samples)
    #
     
    #
    System Benchmarks Index Values               BASELINE       RESULT    INDEX
    #
    Dhrystone 2 using register variables         116700.0   33398098.5   2861.9
    #
    Double-Precision Whetstone                       55.0       5585.1   1015.5
    #
    Execl Throughput                                 43.0       2828.5    657.8
    #
    File Copy 1024 bufsize 2000 maxblocks          3960.0     202957.4    512.5
    #
    File Copy 256 bufsize 500 maxblocks            1655.0      56591.2    341.9
    #
    File Copy 4096 bufsize 8000 maxblocks          5800.0     656227.9   1131.4
    #
    Pipe Throughput                               12440.0     790880.3    635.8
    #
    Pipe-based Context Switching                   4000.0     189209.8    473.0
    #
    Process Creation                                126.0       5694.4    451.9
    #
    Shell Scripts (1 concurrent)                     42.4       5395.5   1272.5
    #
    Shell Scripts (8 concurrent)                      6.0        808.8   1348.0
    #
    System Call Overhead                          15000.0     812367.0    541.6
    #
                                                                       
    ========
    #
    System Benchmarks Index Score                                         776.2 
    PHP Code:
    #
       #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
    #
       #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
    #
       #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
    #
       #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
    #
       #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    #
        ####   #    #  #  #    #          #####   ######  #    #   ####   #    #
    #
     
    #
                     
    4        1           Based on the Byte Magazine Unix Benchmark
    #
                    
    44       11
    #
       
    v   v       4 4        1
    #
        
    v v       44444       1           v4.1 revisions mostly by David CNiemi,
    #
         
    v           4   o   111 WHT.2  RestonVAUSA  <niemi@tux.org>
    #
     
    #
    WHT Variant by Andy ALee <andylee@dslextreme.com>
    #
    Seehttp://www.webhostingtalk.com/showthread.php?s=&threadid=308055
    #
     
    #
     
    #
    Dhrystone 2 using register variables  1 2
    #
     
    #
    Double-Precision Whetstone  1 2
    #
     
    #
    Execl Throughput  1
    #
     
    #
    Filesystem Throughput 1024 bufsize 2000 maxblocks  1
    #
     
    #
    Filesystem Throughput 256 bufsize 500 maxblocks  1
    #
     
    #
    Filesystem Throughput 4096 bufsize 8000 maxblocks  1
    #
     
    #
    Pipe Throughput  1 2
    #
     
    #
    Pipe-based Context Switching  1 2
    #
     
    #
    Process Creation  1
    #
     
    #
    System Call Overhead  1 2
    #
     
    #
    Shell Scripts (8 concurrent)  1
    #
     
    #
    ==============================================================
    #
    BYTE UNIX Benchmarks (Version 4.1-wht.2)
    #
    System -- Linux xen 2.6.26-2-amd64 #1 SMP Wed Jan 13 00:12:41 UTC 2010 x86_64 GNU/Linux
    #
    /dev/mapper/vg0-root  19223252   1055764  17191004   6% /
    #
     
    #
    Start Benchmark RunFri Feb  5 20:07:30 CST 2010
    #
     
    20:07:30 up 15:44,  1 user,  load average0.000.131.03
    #
     
    #
    End Benchmark RunFri Feb  5 20:18:01 CST 2010
    #
     
    20:18:01 up 15:55,  1 user,  load average11.055.022.81
    #
     
    #
     
    #
                         
    INDEX VALUES
    #
    TEST                                        BASELINE     RESULT      INDEX
    #
     
    #
    Dhrystone 2 using register variables        376783.7 12083175.7      320.7
    #
    Double-Precision Whetstone                      83.1     1802.0      216.8
    #
    Execl Throughput                               188.3     2848.7      151.3
    #
    File Copy 1024 bufsize 2000 maxblocks         2672.0   103432.0      387.1
    #
    File Copy 256 bufsize 500 maxblocks           1077.0    27812.0      258.2
    #
    File Read 4096 bufsize 8000 maxblocks        15382.0   877491.0      570.5
    #
    Pipe-based Context Switching                 15448.6   180098.4      116.6
    #
    Pipe Throughput                             111814.6   750025.9       67.1
    #
    Process Creation                               569.3     5624.6       98.8
    #
    Shell Scripts (8 concurrent)                    44.8      819.7      183.0
    #
    System Call Overhead                        114433.5   817131.7       71.4
    #
                                                                     
    =========
    #
         
    FINAL SCORE                                                     179.0 
    UnixBench 5.2 = 776
    WHT UnixBench = 179

    Yes, even on a dual core there is a significant difference.

    Quote Originally Posted by FHDave View Post
    Sirius,

    This thread has been confluted with 135 pages of non-sense. Perhaps it will be better to simply close the thread and start a fresh. Otherwise, the benchmark result of 5.1.2 and 4-wht in the same thread would simply create further confusion.
    Agreed.
    RAM Host -- USA Premium & Budget Linux Hosting
    █ Featuring Powerful cPanel Shared Hosting
    █ & Premium Virtual Dedicated Servers
    Follow us on Twitter
      0 Not allowed!

  5. #2030
    Join Date
    Apr 2009
    Location
    USA / UK
    Posts
    4,577
    Quote Originally Posted by sirius View Post
    Post a link and I'll update the first post.
    http://ramnet.us/dnl/unixbench-5.1.2.tar.gz

    I'll be glad to provide bandwidth for this

    Mine is exactly like the official copy except the Makefile is modified to not build the graphics tests (which causes the compile to fail on most servers):

    GRAPHIC_TESTS = defined

    changed to:

    #GRAPHIC_TESTS = defined

    =====

    Otherwise, if you prefer, the "official" copy:

    http://byte-unixbench.googlecode.com...h-5.1.2.tar.gz

    It requires adjustment to the Makefile to make it compile without the graphics tests for those without X11 installed (and the graphics tests are useless on a server anyway - that's why I made my version with the modified Makefile to make it newbie friendly - make and ./Run just like the old one
      0 Not allowed!

  6. #2031
    Join Date
    Jun 2002
    Location
    PA, USA
    Posts
    5,143
    Please do not start posting modified source for people to download. That's one of the reason why I can't trust all these non-sense coming from the "modified WHT variant." There is much better way to only select the tests you want. Modifying the original source code is not one of them.
    Fluid Hosting, LLC - Enterprise Cloud Infrastructure: Cloud Shared and Reseller, Cloud VPS, and Cloud Hybrid Server
      0 Not allowed!

  7. #2032
    Join Date
    Sep 2008
    Location
    Dallas, TX
    Posts
    4,568
    Quote Originally Posted by FHDave View Post
    Please do not start posting modified source for people to download. That's one of the reason why I can't trust all these non-sense coming from the "modified WHT variant." There is much better way to only select the tests you want. Modifying the original source code is not one of them.
    I don't see what's wrong with what ramnet did, all he did was disable the 3d portion.



    At any rate, mods you should seriously consider closing this thread and starting a new one. Have a g'day chaps.
    Jacob Wall
      0 Not allowed!

  8. #2033
    Join Date
    Jun 2002
    Location
    PA, USA
    Posts
    5,143
    Don't start to make variants and distribute them. Let's stick to the original. That's all I am suggesting. If people start to distribute their own "variants" (just like what we have here in this v4 WHT variant), sooner or later "errors" may start to propagate.

    Besides, there is a better way to disable the graphic test without modifying the code.
    Fluid Hosting, LLC - Enterprise Cloud Infrastructure: Cloud Shared and Reseller, Cloud VPS, and Cloud Hybrid Server
      0 Not allowed!

  9. #2034
    Join Date
    Jun 2002
    Location
    PA, USA
    Posts
    5,143
    I am creating a new thread for the UnixBench 5.1.2 test.

    Let's close this one.
    Fluid Hosting, LLC - Enterprise Cloud Infrastructure: Cloud Shared and Reseller, Cloud VPS, and Cloud Hybrid Server
      0 Not allowed!

  10. #2035
    Join Date
    Apr 2009
    Location
    USA / UK
    Posts
    4,577
    Quote Originally Posted by FHDave View Post
    Don't start to make variants and distribute them. Let's stick to the original. That's all I am suggesting. If people start to distribute their own "variants" (just like what we have here in this v4 WHT variant), sooner or later "errors" may start to propagate.
    There is a huge difference between what I did and what the original OP of this thread did - I modified none of the source, only changed one line in the make file so it will compile on systems without X11 headers.

    The original OP in this thread modified the entire rating scale........basically made a new program.

    In any event, new thread goes far beyond my suggestion of only disabling the graphics tests - I find it ironic how the original OP to this thread weighted disk I/O more heavily, yet you in your new thread have decided to discount disk I/O altogether.......not too sure I agree with that logic very much (disk I/O is an important metric in my opinion).
      0 Not allowed!

  11. #2036
    Join Date
    Jun 2002
    Location
    PA, USA
    Posts
    5,143
    Quote Originally Posted by ramnet View Post
    There is a huge difference between what I did and what the original OP of this thread did - I modified none of the source, only changed one line in the make file so it will compile on systems without X11 headers.

    The original OP in this thread modified the entire rating scale........basically made a new program.

    In any event, new thread goes far beyond my suggestion of only disabling the graphics tests - I find it ironic how the original OP to this thread weighted disk I/O more heavily, yet you in your new thread have decided to discount disk I/O altogether.......not too sure I agree with that logic very much (disk I/O is an important metric in my opinion).

    By disabling the graphic I/O you are also saying that not all test included in the Unixbench package is relevant.

    Anyway ramnet, not trying to argue with you here. My apology if I come out strong. If you don't mind, I will contact you in private regarding the disk I/O test. I have my doubt on this test, really.
    Fluid Hosting, LLC - Enterprise Cloud Infrastructure: Cloud Shared and Reseller, Cloud VPS, and Cloud Hybrid Server
      0 Not allowed!

  12. #2037
    Join Date
    Nov 2007
    Location
    India
    Posts
    843
    dual xeon 5430 with 16GB ram and 6*1.5TB HDD in RAID10

    System Benchmarks Index Score 2927.5

    PHP Code:
    root@server2 [~/unixbench-5.1.2]# ./Run
    make all
    make
    [1]: Entering directory `/root/unixbench-5.1.2'
    Checking distribution of files
    ./pgms  exists
    ./src  exists
    ./testdir  exists
    ./tmp  exists
    ./results  exists
    make[1]: Leaving directory 
    `/root/unixbench-5.1.2'
    sh: 3dinfo: command not found

       #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
       #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
       #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
       #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
       #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
        ####   #    #  #  #    #          #####   ######  #    #   ####   #    #

       Version 5.1.2                      Based on the Byte Magazine Unix Benchmark

       Multi-CPU version                  Version 5 revisions by Ian Smith,
                                          Sunnyvale, CA, USA
       December 22, 2007                  johantheghost at yahoo period com


    1 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

    1 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

    1 x Execl Throughput  1 2 3

    1 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

    1 x File Copy 256 bufsize 500 maxblocks  1 2 3

    1 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

    1 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

    1 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

    1 x Process Creation  1 2 3

    1 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

    1 x Shell Scripts (1 concurrent)  1 2 3

    1 x Shell Scripts (8 concurrent)  1 2 3

    8 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

    8 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

    8 x Execl Throughput  1 2 3

    8 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

    8 x File Copy 256 bufsize 500 maxblocks  1 2 3

    8 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

    8 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

    8 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

    8 x Process Creation  1 2 3

    8 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

    8 x Shell Scripts (1 concurrent)  1 2 3

    8 x Shell Scripts (8 concurrent)  1 2 3

    ========================================================================
       BYTE UNIX Benchmarks (Version 5.1.2)

       System: server2.hostxweb.com: GNU/Linux
       OS: GNU/Linux -- 2.6.18-164.el5 -- #1 SMP Thu Sep 3 03:28:30 EDT 2009
       Machine: x86_64 (x86_64)
       Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
       CPU 0: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (5333.3 bogomips)
              Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
       CPU 1: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (5333.4 bogomips)
              Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
       CPU 2: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (5333.2 bogomips)
              Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
       CPU 3: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (5333.3 bogomips)
              Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
       CPU 4: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (5333.4 bogomips)
              Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
       CPU 5: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (5333.3 bogomips)
              Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
       CPU 6: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (5333.3 bogomips)
              Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
       CPU 7: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (5333.5 bogomips)
              Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
       23:06:38 up 9 days, 16:44,  1 user,  load average: 0.63, 0.50, 0.40; runlevel 3

    ------------------------------------------------------------------------
    Benchmark Run: Fri Feb 05 2010 23:06:38 - 23:35:53
    8 CPUs in system; running 1 parallel copy of tests

    Dhrystone 2 using register variables       14853621.2 lps   (10.0 s, 7 samples)
    Double-Precision Whetstone                     3067.3 MWIPS (9.9 s, 7 samples)
    Execl Throughput                               3211.9 lps   (29.5 s, 2 samples)
    File Copy 1024 bufsize 2000 maxblocks        511012.5 KBps  (30.0 s, 2 samples)
    File Copy 256 bufsize 500 maxblocks          142760.9 KBps  (30.0 s, 2 samples)
    File Copy 4096 bufsize 8000 maxblocks        756265.0 KBps  (30.0 s, 2 samples)
    Pipe Throughput                             1019119.2 lps   (10.0 s, 7 samples)
    Pipe-based Context Switching                 130507.0 lps   (10.0 s, 7 samples)
    Process Creation                              12971.2 lps   (30.0 s, 2 samples)
    Shell Scripts (1 concurrent)                   6430.4 lpm   (60.0 s, 2 samples)
    Shell Scripts (8 concurrent)                   2755.2 lpm   (60.0 s, 2 samples)
    System Call Overhead                         827313.1 lps   (10.0 s, 7 samples)

    System Benchmarks Index Values               BASELINE       RESULT    INDEX
    Dhrystone 2 using register variables         116700.0   14853621.2   1272.8
    Double-Precision Whetstone                       55.0       3067.3    557.7
    Execl Throughput                                 43.0       3211.9    746.9
    File Copy 1024 bufsize 2000 maxblocks          3960.0     511012.5   1290.4
    File Copy 256 bufsize 500 maxblocks            1655.0     142760.9    862.6
    File Copy 4096 bufsize 8000 maxblocks          5800.0     756265.0   1303.9
    Pipe Throughput                               12440.0    1019119.2    819.2
    Pipe-based Context Switching                   4000.0     130507.0    326.3
    Process Creation                                126.0      12971.2   1029.5
    Shell Scripts (1 concurrent)                     42.4       6430.4   1516.6
    Shell Scripts (8 concurrent)                      6.0       2755.2   4592.1
    System Call Overhead                          15000.0     827313.1    551.5
                                                                       ========
    System Benchmarks Index Score                                         982.9

    ------------------------------------------------------------------------
    Benchmark Run: Fri Feb 05 2010 23:35:53 - 00:05:44
    8 CPUs in system; running 8 parallel copies of tests

    Dhrystone 2 using register variables      108810714.1 lps   (10.0 s, 7 samples)
    Double-Precision Whetstone                    24445.9 MWIPS (9.9 s, 7 samples)
    Execl Throughput                              19514.9 lps   (29.4 s, 2 samples)
    File Copy 1024 bufsize 2000 maxblocks        214208.9 KBps  (30.0 s, 2 samples)
    File Copy 256 bufsize 500 maxblocks           54145.8 KBps  (30.0 s, 2 samples)
    File Copy 4096 bufsize 8000 maxblocks        478547.7 KBps  (30.0 s, 2 samples)
    Pipe Throughput                             7628409.9 lps   (10.0 s, 7 samples)
    Pipe-based Context Switching                2342523.0 lps   (10.0 s, 7 samples)
    Process Creation                              64112.8 lps   (30.0 s, 2 samples)
    Shell Scripts (1 concurrent)                  29084.5 lpm   (60.0 s, 2 samples)
    Shell Scripts (8 concurrent)                   3894.0 lpm   (60.0 s, 2 samples)
    System Call Overhead                        2660552.4 lps   (10.0 s, 7 samples)

    System Benchmarks Index Values               BASELINE       RESULT    INDEX
    Dhrystone 2 using register variables         116700.0  108810714.1   9324.0
    Double-Precision Whetstone                       55.0      24445.9   4444.7
    Execl Throughput                                 43.0      19514.9   4538.4
    File Copy 1024 bufsize 2000 maxblocks          3960.0     214208.9    540.9
    File Copy 256 bufsize 500 maxblocks            1655.0      54145.8    327.2
    File Copy 4096 bufsize 8000 maxblocks          5800.0     478547.7    825.1
    Pipe Throughput                               12440.0    7628409.9   6132.2
    Pipe-based Context Switching                   4000.0    2342523.0   5856.3
    Process Creation                                126.0      64112.8   5088.3
    Shell Scripts (1 concurrent)                     42.4      29084.5   6859.6
    Shell Scripts (8 concurrent)                      6.0       3894.0   6490.0
    System Call Overhead                          15000.0    2660552.4   1773.7
                                                                       ========
    System Benchmarks Index Score                                        2927.5

    root@server2 [~/unixbench-5.1.2]# 
    HostNotch Hosting Services 99.9% uptime Shared Hosting, Reseller Hosting
    yajur | Sales Team
    CPanel Hosting • R1 Soft • Offsite-Backup • Great Uptime
    http://hostnotch.com sales @ hostnotch.com
      0 Not allowed!

  13. #2038
    Join Date
    Jun 2002
    Location
    PA, USA
    Posts
    5,143
    Strange, it seems to me on the second run the Disk I/O performance is always much reduced significantly. I have seen this on three out of the four tests so far.
    Fluid Hosting, LLC - Enterprise Cloud Infrastructure: Cloud Shared and Reseller, Cloud VPS, and Cloud Hybrid Server
      0 Not allowed!

  14. #2039
    Join Date
    Jan 2005
    Location
    Jakarta, Indonesia
    Posts
    93
    Xeon E5530, RAM 6 GB DDR3-1066, 4 x 146 GB SAS RAID 5.

    Dhrystone 2 using register variables 10317323.2 lps (10.0 secs, 10 samples)
    Double-Precision Whetstone 2187.8 MWIPS (10.2 secs, 10 samples)
    System Call Overhead 1531374.9 lps (10.0 secs, 10 samples)
    Pipe Throughput 897855.2 lps (10.0 secs, 10 samples)
    Pipe-based Context Switching 159342.0 lps (10.0 secs, 10 samples)
    Process Creation 6670.8 lps (30.0 secs, 3 samples)
    Execl Throughput 2272.3 lps (29.7 secs, 3 samples)
    File Read 1024 bufsize 2000 maxblocks 1753231.0 KBps (30.0 secs, 3 samples)
    File Write 1024 bufsize 2000 maxblocks 851619.0 KBps (30.0 secs, 3 samples)
    File Copy 1024 bufsize 2000 maxblocks 570868.0 KBps (30.0 secs, 3 samples)
    File Read 256 bufsize 500 maxblocks 456095.0 KBps (30.0 secs, 3 samples)
    File Write 256 bufsize 500 maxblocks 232454.0 KBps (30.0 secs, 3 samples)
    File Copy 256 bufsize 500 maxblocks 152603.0 KBps (30.0 secs, 3 samples)
    File Read 4096 bufsize 8000 maxblocks 4314723.0 KBps (30.0 secs, 3 samples)
    File Write 4096 bufsize 8000 maxblocks 2544689.0 KBps (30.0 secs, 3 samples)
    File Copy 4096 bufsize 8000 maxblocks 1584289.0 KBps (30.0 secs, 3 samples)
    Shell Scripts (1 concurrent) 5339.3 lpm (60.0 secs, 3 samples)
    Shell Scripts (8 concurrent) 2300.0 lpm (60.0 secs, 3 samples)
    Shell Scripts (16 concurrent) 1318.3 lpm (60.0 secs, 3 samples)
    Arithmetic Test (type = short) 2260122.5 lps (10.0 secs, 3 samples)
    Arithmetic Test (type = int) 2176413.9 lps (10.0 secs, 3 samples)
    Arithmetic Test (type = long) 2219563.8 lps (10.0 secs, 3 samples)
    Arithmetic Test (type = float) 1028110.1 lps (10.0 secs, 3 samples)
    Arithmetic Test (type = double) 1005370.4 lps (10.0 secs, 3 samples)
    Arithoh 0.0 lps (10.0 secs, 3 samples)
    C Compiler Throughput 1155.0 lpm (60.0 secs, 3 samples)
    Dc: sqrt(2) to 99 decimal places 85447.2 lpm (30.0 secs, 3 samples)
    Recursion Test--Tower of Hanoi 129009.7 lps (20.0 secs, 3 samples)


    INDEX VALUES
    TEST BASELINE RESULT INDEX

    Dhrystone 2 using register variables 116700.0 10317323.2 884.1
    Double-Precision Whetstone 55.0 2187.8 397.8
    Execl Throughput 43.0 2272.3 528.4
    File Copy 1024 bufsize 2000 maxblocks 3960.0 570868.0 1441.6
    File Copy 256 bufsize 500 maxblocks 1655.0 152603.0 922.1
    File Copy 4096 bufsize 8000 maxblocks 5800.0 1584289.0 2731.5
    Pipe Throughput 12440.0 897855.2 721.7
    Process Creation 126.0 6670.8 529.4
    Shell Scripts (8 concurrent) 6.0 2300.0 3833.3
    System Call Overhead 15000.0 1531374.9 1020.9
    =========
    FINAL SCORE 1000.9
      0 Not allowed!

  15. #2040
    Join Date
    Nov 2005
    Posts
    3,944
    I don't see much of a difference between 5.1 and this except for the obvious higher scores in CPU and lower scores in disk. The original unixbench 4 was the same way, the WHT variant put more emphasis on disk to make it more relevant to web hosting.

    It looks like Xen servers are still benchmarking poorly with 5.1 as compared to what OpenVZ/original dedicated servers do. Xen performs poorly in context based switching which is one of the benchmarks done. They still have not figured a way to lower overhead as compared to software based solutions such as openvz.
      0 Not allowed!

Page 82 of 82 FirstFirst ... 327279808182

Posting Permissions

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