The old server benchmark thread in WHT is filled with non-sense results, most, if not all 136 pages of them. Part of the reason is, I believe, the version 4 of UnixBench is not meant to be used on the multiple CPUs/cores environment, pretty much standard on today's servers.
The old Unixbench 4.1-WHT variants produce poor results, especially on Xen environment.
Please do not post any further on that old thread. Do not contribute to further confusion caused by that. IMHO, most, if not all, results on that thread should be discarded.
This thread is meant to be the replacement. Please only post your test using UnixBench 5.1.2. I am posting the original un-modified version. Let's keep this original version in tact.
Simply execute the following command:
tar -zxvf unixbench-5.1.2.tar.gz
cd unixbench-5.1.2
make
./Run
or, if you want to disable file-copy tests, execute the following command (instead of ./Run):
./Run dhry2reg whetstone-double syscall pipe context1 spawn execl shell1 shell8 shell16
Please also note that this 5.1.2 version will make two runs. One run will test only 1 core (1 parallel copy). Another run will test all the available cores (e.g., 16 parallel copies for 8 CPUs with HT).
I am going to post the first "official" result.
Server: Dell PowerEdge R510, Dual E5520 (HyperThreading/Logical CPU DISABLED), 32 GB ECC DDR3 REG,
Code:
BYTE UNIX Benchmarks (Version 5.1.2)
System: localhost.localdomain: GNU/Linux
OS: GNU/Linux -- 2.6.18-164.el5PAE -- #1 SMP Thu Sep 3 04:10:44 EDT 2009
Machine: i686 (i386)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.7 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.6 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 2: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 3: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 4: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 5: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 6: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 7: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
09:59:33 up 3:15, 2 users, load average: 0.10, 0.03, 0.01; runlevel 3
------------------------------------------------------------------------
Benchmark Run: Fri Feb 05 2010 09:59:33 - 10:22:49
8 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 11021090.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2283.3 MWIPS (6.6 s, 7 samples)
Execl Throughput 1826.8 lps (30.0 s, 2 samples)
Pipe Throughput 672125.5 lps (10.0 s, 7 samples)
Pipe-based Context Switching 81365.5 lps (10.0 s, 7 samples)
Process Creation 8635.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4118.9 lpm (60.0 s, 2 samples)
Shell Scripts (16 concurrent) 1180.4 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 2002.9 lpm (60.0 s, 2 samples)
System Call Overhead 2805654.3 lps (10.0 s, 7 samples)
System Benchmarks Partial Index BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 11021090.5 944.4
Double-Precision Whetstone 55.0 2283.3 415.1
Execl Throughput 43.0 1826.8 424.8
Pipe Throughput 12440.0 672125.5 540.3
Pipe-based Context Switching 4000.0 81365.5 203.4
Process Creation 126.0 8635.2 685.3
Shell Scripts (1 concurrent) 42.4 4118.9 971.4
Shell Scripts (16 concurrent) --- 1180.4 ---
Shell Scripts (8 concurrent) 6.0 2002.9 3338.2
System Call Overhead 15000.0 2805654.3 1870.4
========
System Benchmarks Index Score (Partial Only) 751.1
------------------------------------------------------------------------
Benchmark Run: Fri Feb 05 2010 10:22:49 - 10:46:14
8 CPUs in system; running 8 parallel copies of tests
Dhrystone 2 using register variables 87401364.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 17524.5 MWIPS (7.7 s, 7 samples)
Execl Throughput 22554.4 lps (30.0 s, 2 samples)
Pipe Throughput 5382685.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1849908.9 lps (10.0 s, 7 samples)
Process Creation 86373.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 23144.0 lpm (60.0 s, 2 samples)
Shell Scripts (16 concurrent) 2405.8 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 4812.1 lpm (60.1 s, 2 samples)
System Call Overhead 5699670.5 lps (10.0 s, 7 samples)
System Benchmarks Partial Index BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 87401364.4 7489.4
Double-Precision Whetstone 55.0 17524.5 3186.3
Execl Throughput 43.0 22554.4 5245.2
Pipe Throughput 12440.0 5382685.1 4326.9
Pipe-based Context Switching 4000.0 1849908.9 4624.8
Process Creation 126.0 86373.6 6855.0
Shell Scripts (1 concurrent) 42.4 23144.0 5458.5
Shell Scripts (16 concurrent) --- 2405.8 ---
Shell Scripts (8 concurrent) 6.0 4812.1 8020.2
System Call Overhead 15000.0 5699670.5 3799.8
========
System Benchmarks Index Score (Partial Only) 5215.7
Same server, with HyperThreading/Logical CPUs ENABLEDCode:
BYTE UNIX Benchmarks (Version 5.1.2)
System: localhost.localdomain: GNU/Linux
OS: GNU/Linux -- 2.6.18-164.11.1.el5PAE -- #1 SMP Wed Jan 20 08:16:13 EST 2010
Machine: i686 (i386)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.6 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 2: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 3: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 4: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 5: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 6: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 7: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 8: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 9: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.6 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 10: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 11: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.6 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 12: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 13: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 14: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 15: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.6 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
11:32:24 up 2 min, 2 users, load average: 0.09, 0.08, 0.03; runlevel 3
------------------------------------------------------------------------
Benchmark Run: Fri Feb 05 2010 11:32:24 - 11:55:40
16 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 10898006.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2265.2 MWIPS (6.6 s, 7 samples)
Execl Throughput 1836.7 lps (30.0 s, 2 samples)
Pipe Throughput 675419.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching 84729.4 lps (10.0 s, 7 samples)
Process Creation 8720.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4076.6 lpm (60.0 s, 2 samples)
Shell Scripts (16 concurrent) 1370.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 2160.2 lpm (60.0 s, 2 samples)
System Call Overhead 2759595.6 lps (10.0 s, 7 samples)
System Benchmarks Partial Index BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 10898006.5 933.8
Double-Precision Whetstone 55.0 2265.2 411.9
Execl Throughput 43.0 1836.7 427.1
Pipe Throughput 12440.0 675419.8 542.9
Pipe-based Context Switching 4000.0 84729.4 211.8
Process Creation 126.0 8720.0 692.1
Shell Scripts (1 concurrent) 42.4 4076.6 961.5
Shell Scripts (16 concurrent) --- 1370.5 ---
Shell Scripts (8 concurrent) 6.0 2160.2 3600.3
System Call Overhead 15000.0 2759595.6 1839.7
========
System Benchmarks Index Score (Partial Only) 758.7
------------------------------------------------------------------------
Benchmark Run: Fri Feb 05 2010 11:55:40 - 12:19:15
16 CPUs in system; running 16 parallel copies of tests
Dhrystone 2 using register variables 95169178.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone 29861.0 MWIPS (8.3 s, 7 samples)
Execl Throughput 33699.1 lps (30.0 s, 2 samples)
Pipe Throughput 7363181.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 2543525.1 lps (10.0 s, 7 samples)
Process Creation 130941.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 35122.8 lpm (60.0 s, 2 samples)
Shell Scripts (16 concurrent) 3302.9 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 6429.4 lpm (60.0 s, 2 samples)
System Call Overhead 6630712.0 lps (10.0 s, 7 samples)
System Benchmarks Partial Index BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 95169178.1 8155.0
Double-Precision Whetstone 55.0 29861.0 5429.3
Execl Throughput 43.0 33699.1 7837.0
Pipe Throughput 12440.0 7363181.0 5919.0
Pipe-based Context Switching 4000.0 2543525.1 6358.8
Process Creation 126.0 130941.4 10392.2
Shell Scripts (1 concurrent) 42.4 35122.8 8283.7
Shell Scripts (16 concurrent) --- 3302.9 ---
Shell Scripts (8 concurrent) 6.0 6429.4 10715.6
System Call Overhead 15000.0 6630712.0 4420.5
========
System Benchmarks Index Score (Partial Only) 7219.1
~38.4% improvement with HT.Post yours. Again, please disable disk I/O test. Use the command above.
PS: Thanks to John (JTY) for pointing me towards testing using the 5.1.2 version!