Solid-Run　Cubox-i4にUbuntuをインストール(その3）

あまりにも遅いから、UnixBenchで速さを調べてみます。

cd /usr/local/src
$ wget http://byte-unixbench.googlecode.com/files/UnixBench5.1.3.tgz
$ tar xzf UnixBench5.1.3.tgz
$ cd UnixBench5.1.3
$ make

Index Score 216

うーん、いいのか悪いのかわからない。詳細スコアはちょっと待ってください。

参考までに、RaspberryPiのScoreは103ぐらいとのこと。Cubox-i4はコアが4倍あるはずなので、伸びは悪いですね。

ベンチマークの取り直し

UnixBenchがどうも正しくＣＰＵの数を認識してくれないようなので、

./Run -c 4

強制的に4並列で計算させた結果の例がこちらです。Index Score 538は、結構いいですね。でも体感として絶対にこんなに速くありませんから、まだいろいろ問題がありそうです。*1lightdm上で実行しています。

Benchmark of linaro-ubuntu-desktop / GNU/Linux on Sun Dec 08 1929

BYTE UNIX Benchmarks (Version 5.1.3)

Test System Information

System:	linaro-ubuntu-desktop: GNU/Linux
OS:	GNU/Linux -- 3.0.35-gbe699c6-dirty -- #11 SMP PREEMPT Tue Jan 21 20:51:32 IST 2014
Machine:	armv7l: armv7l
Language:	en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
Uptime:	01:03:44 up 7 days, 4:15, 4 users, load average: 6.22, 5.71, 5.57; runlevel 2

Benchmark Run: 0 CPUs; 4 parallel processes

Time: 01:03:44 - 01:33:41; 29m 57s

System Benchmarks

Test	Score	Unit	Time	Iters.	Baseline	Index
Dhrystone 2 using register variables	`8785262.5`	`lps`	`10.0 s`	`7`	`116700.0`	`752.8`
Double-Precision Whetstone	`2761.9`	`MWIPS`	`9.9 s`	`7`	`55.0`	`502.2`
Execl Throughput	`1836.3`	`lps`	`29.6 s`	`2`	`43.0`	`427.0`
File Copy 1024 bufsize 2000 maxblocks	`144808.4`	`KBps`	`30.0 s`	`2`	`3960.0`	`365.7`
File Copy 256 bufsize 500 maxblocks	`39705.0`	`KBps`	`30.0 s`	`2`	`1655.0`	`239.9`
File Copy 4096 bufsize 8000 maxblocks	`342599.9`	`KBps`	`30.0 s`	`2`	`5800.0`	`590.7`
Pipe Throughput	`844052.0`	`lps`	`10.0 s`	`7`	`12440.0`	`678.5`
Pipe-based Context Switching	`226979.1`	`lps`	`10.0 s`	`7`	`4000.0`	`567.4`
Process Creation	`4194.8`	`lps`	`30.0 s`	`2`	`126.0`	`332.9`
Shell Scripts (1 concurrent)	`2893.1`	`lpm`	`60.1 s`	`2`	`42.4`	`682.3`
Shell Scripts (8 concurrent)	`468.2`	`lpm`	`60.3 s`	`2`	`6.0`	`780.3`
System Call Overhead	`1552269.8`	`lps`	`10.0 s`	`7`	`15000.0`	`1034.8`
System Benchmarks Index Score:						`538.2`

No Warranties: This information is provided free of charge and "as is" without any warranty, condition, or representation of any kind, either express or implied, including but not limited to, any warranty respecting non-infringement, and the implied warranties of conditions of merchantability and fitness for a particular purpose. All logos or trademarks on this site are the property of their respective owner. In no event shall the author be liable for any direct, indirect, special, incidental, consequential or other damages howsoever caused whether arising in contract, tort, or otherwise, arising out of or in connection with the use or performance of the information contained on this web site.

SDカードの変更

当初使っていたものが、SanDiskの8GByte microSD HC class4。

手元にある一番いいスペックのものを持ってきて、SanDiskの32GByte SDHC UHS-I class 10 Max 30MB/Sec。

変に動作が突っかかるような挙動が無くなり、そこそこ速くなりました。ちゃんと良いmicroSDカードを選んで使ったほうが良さそうです。

Benchmark of linaro-ubuntu-desktop / GNU/Linux on Sun Dec 08 1929

BYTE UNIX Benchmarks (Version 5.1.3)

Test System Information

System:	linaro-ubuntu-desktop: GNU/Linux
OS:	GNU/Linux -- 3.0.35-gbe699c6-dirty -- #11 SMP PREEMPT Tue Jan 21 20:51:32 IST 2014
Machine:	armv7l: armv7l
Language:	en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
Uptime:	03:31:39 up 3 min, 2 users, load average: 4.79, 2.61, 1.05; runlevel 2

Benchmark Run: 0 CPUs; 4 parallel processes

Time: 03:31:39 - 04:01:09; 29m 30s

System Benchmarks

Test	Score	Unit	Time	Iters.	Baseline	Index
Dhrystone 2 using register variables	`12051825.8`	`lps`	`10.0 s`	`7`	`116700.0`	`1032.7`
Double-Precision Whetstone	`2895.8`	`MWIPS`	`10.0 s`	`7`	`55.0`	`526.5`
Execl Throughput	`2315.5`	`lps`	`29.7 s`	`2`	`43.0`	`538.5`
File Copy 1024 bufsize 2000 maxblocks	`126381.7`	`KBps`	`30.0 s`	`2`	`3960.0`	`319.1`
File Copy 256 bufsize 500 maxblocks	`35185.0`	`KBps`	`30.0 s`	`2`	`1655.0`	`212.6`
File Copy 4096 bufsize 8000 maxblocks	`329713.5`	`KBps`	`30.0 s`	`2`	`5800.0`	`568.5`
Pipe Throughput	`578458.3`	`lps`	`10.0 s`	`7`	`12440.0`	`465.0`
Pipe-based Context Switching	`155220.0`	`lps`	`10.0 s`	`7`	`4000.0`	`388.0`
Process Creation	`5329.3`	`lps`	`30.0 s`	`2`	`126.0`	`423.0`
Shell Scripts (1 concurrent)	`3702.0`	`lpm`	`60.0 s`	`2`	`42.4`	`873.1`
Shell Scripts (8 concurrent)	`481.2`	`lpm`	`60.3 s`	`2`	`6.0`	`802.1`
System Call Overhead	`2161972.6`	`lps`	`10.0 s`	`7`	`15000.0`	`1441.3`
System Benchmarks Index Score:						`555.9`