Wie versprochen, hier ein Benchmark vom RS 8000 in der SSD-Variante:
Entweder lese ich den Benchmark verkehrt oder der RS 8000 ist unterm Strich langsamer als ein EX4 mit Festplatte (inzwischen nicht mehr im Programm) eines Mitbewerbers:
Code
BYTE UNIX Benchmarks (Version 5.1.3)
System: srv1: GNU/Linux
OS: GNU/Linux -- 2.6.32.59-grsec -- #3 SMP Thu May 10 09:17:28 CEST 2012
Machine: x86_64 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (6824.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (6822.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 2: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (6822.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 3: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (6822.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 4: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (6822.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 5: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (6822.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 6: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (6822.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 7: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (6822.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
17:21:03 up 427 days, 21:54, 1 user, load average: 1,18, 0,80, 0,59; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Mo Mai 23 2016 17:21:03 - 17:50:34
8 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 43427246.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4251.5 MWIPS (11.1 s, 7 samples)
Execl Throughput 3380.1 lps (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1506357.6 KBps (30.1 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 428914.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2913393.0 KBps (30.0 s, 2 samples)
Pipe Throughput 3012926.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 167720.5 lps (10.0 s, 7 samples)
Process Creation 12801.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 6293.0 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 4370.5 lpm (60.0 s, 2 samples)
System Call Overhead 4814086.2 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 43427246.8 3721.3
Double-Precision Whetstone 55.0 4251.5 773.0
Execl Throughput 43.0 3380.1 786.1
File Copy 1024 bufsize 2000 maxblocks 3960.0 1506357.6 3803.9
File Copy 256 bufsize 500 maxblocks 1655.0 428914.3 2591.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 2913393.0 5023.1
Pipe Throughput 12440.0 3012926.4 2422.0
Pipe-based Context Switching 4000.0 167720.5 419.3
Process Creation 126.0 12801.5 1016.0
Shell Scripts (1 concurrent) 42.4 6293.0 1484.2
Shell Scripts (8 concurrent) 6.0 4370.5 7284.2
System Call Overhead 15000.0 4814086.2 3209.4
========
System Benchmarks Index Score 1996.4
------------------------------------------------------------------------
Benchmark Run: Mo Mai 23 2016 17:50:34 - 18:22:17
8 CPUs in system; running 8 parallel copies of tests
Dhrystone 2 using register variables 162830877.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 30802.3 MWIPS (9.9 s, 7 samples)
Execl Throughput 31834.9 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1176153.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 340081.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 7943310.6 KBps (30.0 s, 2 samples)
Pipe Throughput 12440069.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 2897152.7 lps (10.0 s, 7 samples)
Process Creation 102977.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 39741.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 5421.5 lpm (60.1 s, 2 samples)
System Call Overhead 10649284.1 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 162830877.4 13952.9
Double-Precision Whetstone 55.0 30802.3 5600.4
Execl Throughput 43.0 31834.9 7403.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 1176153.7 2970.1
File Copy 256 bufsize 500 maxblocks 1655.0 340081.3 2054.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 7943310.6 13695.4
Pipe Throughput 12440.0 12440069.6 10000.1
Pipe-based Context Switching 4000.0 2897152.7 7242.9
Process Creation 126.0 102977.8 8172.8
Shell Scripts (1 concurrent) 42.4 39741.5 9373.0
Shell Scripts (8 concurrent) 6.0 5421.5 9035.8
System Call Overhead 15000.0 10649284.1 7099.5
========
System Benchmarks Index Score 7128.3
Display More
Bei der Festplatte ist er natürlich langsamer:
Code
hdparm -tT /dev/sda
/dev/sda:
Timing cached reads: 23934 MB in 2.00 seconds = 11979.33 MB/sec
Timing buffered disk reads: 440 MB in 3.01 seconds = 146.12 MB/sec
hdparm -tT --direct /dev/sda
/dev/sda:
Timing O_DIRECT cached reads: 562 MB in 2.00 seconds = 280.66 MB/sec
Timing O_DIRECT disk reads: 314 MB in 3.01 seconds = 104.33 MB/sec
Display More