自宅サーバマザーボード交換（GIGABYTE:GA-C1007UN-D） CentOS6.5
先日、サーバのファンを換えたときに久しぶりに中をみた。
特段悪いところがあるわけではないのだけど、もう４年も使っているようで…。
個体コンデンサで固めたGIGABYTEのマザーボードはまだまだ稼働できそうだし、Atom330に不満を感じているわけでもない。
しかし、ここらで気分も変えつつ、予防保全の意味も兼ねてマザーボードを換えることにした。
Atom330の後釜をどうしようか難しかったのだけど、Celeron 1007Uに決定。オンボードでこれを搭載したマザーボード「GIGABYTE マザーボード Intel NM70 Celeron 1007U搭載 Mini ITX GA-C1007UN-D
OSもCentOS6.5に変更。
メモリもこれまでDDR2だったので、新しいボードに適合するDDR3の2G(Transcend デスクトップPC用 PC3-10600(DDR3-1333) 2GB 1.5V 240pin DIMM (無期限保証) JM1333KLN-2G
せっかくなので性能に変化があるか見たかったので、UnixBenchでベンチマークを行ってみた。
■構成変更前
・マザーボード：GIGABYTE GA-GC330UD（Atom330搭載）
・メモリ：バルク1Gbyte
BYTE UNIX Benchmarks (Version 5.1.3)
System: silvia: GNU/Linux
OS: GNU/Linux — 2.6.18-371.1.2.el5 — #1 SMP Tue Oct 22 12:57:43 EDT 2013
Machine: i686 (i386)
Language: en_US.utf8 (charmap=”UTF-8″, collate=”UTF-8″)
CPU 0: Intel(R) Atom(TM) CPU 330 @ 1.60GHz (3200.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext
CPU 1: Intel(R) Atom(TM) CPU 330 @ 1.60GHz (3199.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext
CPU 2: Intel(R) Atom(TM) CPU 330 @ 1.60GHz (3200.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext
CPU 3: Intel(R) Atom(TM) CPU 330 @ 1.60GHz (3200.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext
13:51:41 up 6 days, 10:20, 1 user, load average: 0.12, 0.03, 0.01; runlevel 3
————————————————————————
Benchmark Run: 日 12月 01 2013 13:51:41 – 14:19:56
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 2906910.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 579.9 MWIPS (10.1 s, 7 samples)
Execl Throughput 1512.4 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 131536.9 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 37049.6 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 359763.4 KBps (30.0 s, 2 samples)
Pipe Throughput 308199.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 50970.2 lps (10.0 s, 7 samples)
Process Creation 5967.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2167.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 583.2 lpm (60.0 s, 2 samples)
System Call Overhead 448609.9 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 2906910.3 249.1
Double-Precision Whetstone 55.0 579.9 105.4
Execl Throughput 43.0 1512.4 351.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 131536.9 332.2
File Copy 256 bufsize 500 maxblocks 1655.0 37049.6 223.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 359763.4 620.3
Pipe Throughput 12440.0 308199.1 247.7
Pipe-based Context Switching 4000.0 50970.2 127.4
Process Creation 126.0 5967.2 473.6
Shell Scripts (1 concurrent) 42.4 2167.9 511.3
Shell Scripts (8 concurrent) 6.0 583.2 972.1
System Call Overhead 15000.0 448609.9 299.1
========
System Benchmarks Index Score 314.8
————————————————————————
Benchmark Run: 日 12月 01 2013 14:19:56 – 14:48:22
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 7805478.6 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2063.8 MWIPS (10.1 s, 7 samples)
Execl Throughput 3903.7 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 107519.2 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 23856.7 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 294943.3 KBps (30.1 s, 2 samples)
Pipe Throughput 602267.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 197420.5 lps (10.0 s, 7 samples)
Process Creation 12702.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4473.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 650.4 lpm (60.1 s, 2 samples)
System Call Overhead 1009038.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 7805478.6 668.8
Double-Precision Whetstone 55.0 2063.8 375.2
Execl Throughput 43.0 3903.7 907.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 107519.2 271.5
File Copy 256 bufsize 500 maxblocks 1655.0 23856.7 144.1
File Copy 4096 bufsize 8000 maxblocks 5800.0 294943.3 508.5
Pipe Throughput 12440.0 602267.6 484.1
Pipe-based Context Switching 4000.0 197420.5 493.6
Process Creation 126.0 12702.2 1008.1
Shell Scripts (1 concurrent) 42.4 4473.6 1055.1
Shell Scripts (8 concurrent) 6.0 650.4 1083.9
System Call Overhead 15000.0 1009038.4 672.7
========
System Benchmarks Index Score 554.2
■構成変更後
・マザーボード：GIGABYTE GA-C1007UN-D（Celeron 1007U搭載）
・メモリ：Transcend JM1333KLN-1Gbyte
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: SILVIA-SERVER: GNU/Linux
OS: GNU/Linux — 2.6.32-431.el6.i686 — #1 SMP Fri Nov 22 00:26:36 UTC 2013
Machine: i686 (i386)
Language: en_US.utf8 (charmap=”UTF-8″, collate=”UTF-8″)
CPU 0: Intel(R) Celeron(R) CPU 1007U @ 1.50GHz (2993.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
CPU 1: Intel(R) Celeron(R) CPU 1007U @ 1.50GHz (2993.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, Intel virtualization
00:54:53 up 4:41, 1 user, load average: 0.70, 0.26, 0.09; runlevel 3
————————————————————————
Benchmark Run: Mon Dec 02 2013 00:54:54 – 01:24:44
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 12400810.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 1481.3 MWIPS (10.2 s, 7 samples)
Execl Throughput 2768.8 lps (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 292713.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 79011.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 859602.9 KBps (30.0 s, 2 samples)
Pipe Throughput 639571.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 150262.8 lps (10.0 s, 7 samples)
Process Creation 7479.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2678.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 410.4 lpm (60.1 s, 2 samples)
System Call Overhead 982500.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 12400810.3 1062.6
Double-Precision Whetstone 55.0 1481.3 269.3
Execl Throughput 43.0 2768.8 643.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 292713.3 739.2
File Copy 256 bufsize 500 maxblocks 1655.0 79011.4 477.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 859602.9 1482.1
Pipe Throughput 12440.0 639571.9 514.1
Pipe-based Context Switching 4000.0 150262.8 375.7
Process Creation 126.0 7479.4 593.6
Shell Scripts (1 concurrent) 42.4 2678.6 631.8
Shell Scripts (8 concurrent) 6.0 410.4 684.0
System Call Overhead 15000.0 982500.5 655.0
========
System Benchmarks Index Score 618.4
————————————————————————
Benchmark Run: Mon Dec 02 2013 01:24:44 – 01:55:54
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 13589710.6 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2962.9 MWIPS (10.2 s, 7 samples)
Execl Throughput 3807.7 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 355965.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 90430.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1235659.5 KBps (30.0 s, 2 samples)
Pipe Throughput 648347.2 lps (10.0 s, 7 samples)
Pipe-based Context Switching 150901.7 lps (10.0 s, 7 samples)
Process Creation 8736.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2988.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 440.1 lpm (60.1 s, 2 samples)
System Call Overhead 1033292.8 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 13589710.6 1164.5
Double-Precision Whetstone 55.0 2962.9 538.7
Execl Throughput 43.0 3807.7 885.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 355965.0 898.9
File Copy 256 bufsize 500 maxblocks 1655.0 90430.8 546.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 1235659.5 2130.4
Pipe Throughput 12440.0 648347.2 521.2
Pipe-based Context Switching 4000.0 150901.7 377.3
Process Creation 126.0 8736.8 693.4
Shell Scripts (1 concurrent) 42.4 2988.4 704.8
Shell Scripts (8 concurrent) 6.0 440.1 733.6
System Call Overhead 15000.0 1033292.8 688.9
========
System Benchmarks Index Score 742.8
スコアは上がったようです。消費電力については、CPUのTDPは上がってしまっているのだけど、サーバ全体で同じくらいか下がっているといいのだけど。これについてはワットチェッカーとか使っていないので未確認。