ITmedia NEWS > STUDIO >
連載
» 2020年11月28日 07時00分 公開

名刺サイズの超小型PC「ラズパイ」で遊ぶ(第33回):ラズパイ一体型キーボード「Raspberry Pi 400」の性能は? ベンチマークテストで従来モデルと比べた

かつての“マイコン”を思わせるラズパイ一体型キーボード「Raspberry Pi 400」でベンチマークテストを実施。その実力とは?

[岩泉茂,ITmedia]

 前回はラズパイ一体型キーボード「Raspberry Pi 400」の外観をご紹介した他、「技術基準適合証明」(技適)を通っていないデバイスを試すときの手続きについて紹介しました。今回はベンチマークソフト「UnixBench」を使って実機のパフォーマンスを確認していきます。

 UnixBenchはLinuxでは定番となっているベンチマークテストで、整数処理演算や浮動小数点演算、ファイルのコピーなどを行ってシステム全体のパフォーマンスを計測します。最終的には「System Benchmarks Index Score」の値で総合スコアが表示されますが、これは「Sun SPARCstation 20 SM61」という、1995年頃のコンピュータの性能を10として比較したときの値です。

 UnixBenchのインストールは、GitHubにアップされているファイルをフォルダごとそのままダウンロードすればOKです。UnixBenchがインストールされたディレクトリに移動して「./Run」で実行します。

$ git clone https://github.com/kdlucas/byte-unixbench
$ cd byte-unixbench/UnixBench
$ ./Run

 今回はRaspberry Pi 400とRaspberry Pi 4の2GBモデル、4GBモデル、8GBモデルの4つでUnixBenchを使ったベンチマークテストを行いました。それぞれでシングルコア、マルチコアでのテストとなります。結果は以下の通りです。

UnixBench UnixBenchでのテスト結果(クリックで拡大)

 Raspberry Pi 4よりもRaspberry Pi 400がかなりいい値を示しています。これは搭載されているSoCの違いです。

 Raspberry Pi 4はSoCが「Broadcom BCM2711(Quad core Cortex-A72 (ARM v8) 64-bit SoC @ 1.5GHz」だったのに対して、Raspberry Pi 400は駆動周波数が高い「Broadcom BCM2711C0(quad-core Cortex-A72 (ARM v8) 64-bit SoC @ 1.8GHz」を搭載しているためです。

 またUnixBenchはCPUの性能が高いほどいい結果となるので、これだけの差がついたのでしょう。たったの0.3GHzですが、随分と違うのだなあと思いました。

 ちなみにRaspberry Pi 4は搭載メモリの容量による性能の差はありません。

 今回のテスト結果について、以下に記しておきます。

Raspberry Pi 4 2GB

------------------------------------------------------------------------
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables       10207716.0 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2387.5 MWIPS (9.6 s, 7 samples)
Execl Throughput                                806.6 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks         95729.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           26300.6 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        269064.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                              106029.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  39209.8 lps   (10.0 s, 7 samples)
Process Creation                               1570.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2368.5 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    695.6 lpm   (60.0 s, 2 samples)
System Call Overhead                         489205.5 lps   (10.0 s, 7 samples)
System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   10207716.0    874.7
Double-Precision Whetstone                       55.0       2387.5    434.1
Execl Throughput                                 43.0        806.6    187.6
File Copy 1024 bufsize 2000 maxblocks          3960.0      95729.5    241.7
File Copy 256 bufsize 500 maxblocks            1655.0      26300.6    158.9
File Copy 4096 bufsize 8000 maxblocks          5800.0     269064.9    463.9
Pipe Throughput                               12440.0     106029.8     85.2
Pipe-based Context Switching                   4000.0      39209.8     98.0
Process Creation                                126.0       1570.1    124.6
Shell Scripts (1 concurrent)                     42.4       2368.5    558.6
Shell Scripts (8 concurrent)                      6.0        695.6   1159.3
System Call Overhead                          15000.0     489205.5    326.1
                                                                   ========
System Benchmarks Index Score                                         284.3
------------------------------------------------------------------------
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables       39533874.0 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     9408.2 MWIPS (9.4 s, 7 samples)
Execl Throughput                               2455.3 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        183966.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           50218.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        565043.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              398553.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 155854.0 lps   (10.0 s, 7 samples)
Process Creation                               4115.9 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   5563.4 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    755.7 lpm   (60.2 s, 2 samples)
System Call Overhead                        1855972.2 lps   (10.0 s, 7 samples)
System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   39533874.0   3387.6
Double-Precision Whetstone                       55.0       9408.2   1710.6
Execl Throughput                                 43.0       2455.3    571.0
File Copy 1024 bufsize 2000 maxblocks          3960.0     183966.5    464.6
File Copy 256 bufsize 500 maxblocks            1655.0      50218.0    303.4
File Copy 4096 bufsize 8000 maxblocks          5800.0     565043.5    974.2
Pipe Throughput                               12440.0     398553.7    320.4
Pipe-based Context Switching                   4000.0     155854.0    389.6
Process Creation                                126.0       4115.9    326.7
Shell Scripts (1 concurrent)                     42.4       5563.4   1312.1
Shell Scripts (8 concurrent)                      6.0        755.7   1259.5
System Call Overhead                          15000.0    1855972.2   1237.3
                                                                   ========
System Benchmarks Index Score                                         761.3
------------------------------------------------------------------------

Raspberry Pi 4 4GB

------------------------------------------------------------------------
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables       10120286.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2387.9 MWIPS (9.6 s, 7 samples)
Execl Throughput                                807.7 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks         95632.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           26242.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        270872.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                              106599.9 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  37916.4 lps   (10.0 s, 7 samples)
Process Creation                               1578.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2487.1 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    707.2 lpm   (60.1 s, 2 samples)
System Call Overhead                         490917.6 lps   (10.0 s, 7 samples)
System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   10120286.5    867.2
Double-Precision Whetstone                       55.0       2387.9    434.2
Execl Throughput                                 43.0        807.7    187.8
File Copy 1024 bufsize 2000 maxblocks          3960.0      95632.4    241.5
File Copy 256 bufsize 500 maxblocks            1655.0      26242.5    158.6
File Copy 4096 bufsize 8000 maxblocks          5800.0     270872.0    467.0
Pipe Throughput                               12440.0     106599.9     85.7
Pipe-based Context Switching                   4000.0      37916.4     94.8
Process Creation                                126.0       1578.8    125.3
Shell Scripts (1 concurrent)                     42.4       2487.1    586.6
Shell Scripts (8 concurrent)                      6.0        707.2   1178.7
System Call Overhead                          15000.0     490917.6    327.3
                                                                   ========
System Benchmarks Index Score                                         285.3
------------------------------------------------------------------------
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables       39731765.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     9459.8 MWIPS (9.4 s, 7 samples)
Execl Throughput                               2470.9 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        183548.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           50223.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        565826.2 KBps  (30.0 s, 2 samples)
Pipe Throughput                              401295.3 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 149653.4 lps   (10.0 s, 7 samples)
Process Creation                               4187.3 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   5589.3 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    763.7 lpm   (60.1 s, 2 samples)
System Call Overhead                        1810370.7 lps   (10.0 s, 7 samples)
System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   39731765.6   3404.6
Double-Precision Whetstone                       55.0       9459.8   1720.0
Execl Throughput                                 43.0       2470.9    574.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     183548.5    463.5
File Copy 256 bufsize 500 maxblocks            1655.0      50223.3    303.5
File Copy 4096 bufsize 8000 maxblocks          5800.0     565826.2    975.6
Pipe Throughput                               12440.0     401295.3    322.6
Pipe-based Context Switching                   4000.0     149653.4    374.1
Process Creation                                126.0       4187.3    332.3
Shell Scripts (1 concurrent)                     42.4       5589.3   1318.2
Shell Scripts (8 concurrent)                      6.0        763.7   1272.8
System Call Overhead                          15000.0    1810370.7   1206.9
                                                                   ========
System Benchmarks Index Score                                         760.6
------------------------------------------------------------------------

Raspberry Pi 4 8GB

------------------------------------------------------------------------
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables       10161367.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2387.3 MWIPS (9.6 s, 7 samples)
Execl Throughput                                808.3 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks         92372.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           25185.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        264635.4 KBps  (30.0 s, 2 samples)
Pipe Throughput                              105707.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  38391.1 lps   (10.0 s, 7 samples)
Process Creation                               1562.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2340.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    710.1 lpm   (60.1 s, 2 samples)
System Call Overhead                         490097.4 lps   (10.0 s, 7 samples)
System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   10161367.6    870.7
Double-Precision Whetstone                       55.0       2387.3    434.1
Execl Throughput                                 43.0        808.3    188.0
File Copy 1024 bufsize 2000 maxblocks          3960.0      92372.0    233.3
File Copy 256 bufsize 500 maxblocks            1655.0      25185.0    152.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     264635.4    456.3
Pipe Throughput                               12440.0     105707.1     85.0
Pipe-based Context Switching                   4000.0      38391.1     96.0
Process Creation                                126.0       1562.8    124.0
Shell Scripts (1 concurrent)                     42.4       2340.2    551.9
Shell Scripts (8 concurrent)                      6.0        710.1   1183.4
System Call Overhead                          15000.0     490097.4    326.7
                                                                   ========
System Benchmarks Index Score                                         281.6
------------------------------------------------------------------------
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables       40512509.4 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     9542.7 MWIPS (9.6 s, 7 samples)
Execl Throughput                               2484.5 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        181985.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           50011.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        565057.1 KBps  (30.0 s, 2 samples)
Pipe Throughput                              408483.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 158037.5 lps   (10.0 s, 7 samples)
Process Creation                               4175.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   5677.1 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    770.9 lpm   (60.2 s, 2 samples)
System Call Overhead                        1860702.6 lps   (10.0 s, 7 samples)
System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   40512509.4   3471.5
Double-Precision Whetstone                       55.0       9542.7   1735.0
Execl Throughput                                 43.0       2484.5    577.8
File Copy 1024 bufsize 2000 maxblocks          3960.0     181985.8    459.6
File Copy 256 bufsize 500 maxblocks            1655.0      50011.0    302.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     565057.1    974.2
Pipe Throughput                               12440.0     408483.8    328.4
Pipe-based Context Switching                   4000.0     158037.5    395.1
Process Creation                                126.0       4175.1    331.4
Shell Scripts (1 concurrent)                     42.4       5677.1   1338.9
Shell Scripts (8 concurrent)                      6.0        770.9   1284.8
System Call Overhead                          15000.0    1860702.6   1240.5
                                                                   ========
System Benchmarks Index Score                                         769.6
------------------------------------------------------------------------

Raspberry Pi 400

------------------------------------------------------------------------
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables       10161367.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2387.3 MWIPS (9.6 s, 7 samples)
Execl Throughput                                808.3 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks         92372.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           25185.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        264635.4 KBps  (30.0 s, 2 samples)
Pipe Throughput                              105707.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  38391.1 lps   (10.0 s, 7 samples)
Process Creation                               1562.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2340.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    710.1 lpm   (60.1 s, 2 samples)
System Call Overhead                         490097.4 lps   (10.0 s, 7 samples)
System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   10161367.6    870.7
Double-Precision Whetstone                       55.0       2387.3    434.1
Execl Throughput                                 43.0        808.3    188.0
File Copy 1024 bufsize 2000 maxblocks          3960.0      92372.0    233.3
File Copy 256 bufsize 500 maxblocks            1655.0      25185.0    152.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     264635.4    456.3
Pipe Throughput                               12440.0     105707.1     85.0
Pipe-based Context Switching                   4000.0      38391.1     96.0
Process Creation                                126.0       1562.8    124.0
Shell Scripts (1 concurrent)                     42.4       2340.2    551.9
Shell Scripts (8 concurrent)                      6.0        710.1   1183.4
System Call Overhead                          15000.0     490097.4    326.7
                                                                   ========
System Benchmarks Index Score                                         281.6
------------------------------------------------------------------------
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables       40512509.4 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     9542.7 MWIPS (9.6 s, 7 samples)
Execl Throughput                               2484.5 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        181985.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           50011.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        565057.1 KBps  (30.0 s, 2 samples)
Pipe Throughput                              408483.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 158037.5 lps   (10.0 s, 7 samples)
Process Creation                               4175.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   5677.1 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    770.9 lpm   (60.2 s, 2 samples)
System Call Overhead                        1860702.6 lps   (10.0 s, 7 samples)
System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   40512509.4   3471.5
Double-Precision Whetstone                       55.0       9542.7   1735.0
Execl Throughput                                 43.0       2484.5    577.8
File Copy 1024 bufsize 2000 maxblocks          3960.0     181985.8    459.6
File Copy 256 bufsize 500 maxblocks            1655.0      50011.0    302.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     565057.1    974.2
Pipe Throughput                               12440.0     408483.8    328.4
Pipe-based Context Switching                   4000.0     158037.5    395.1
Process Creation                                126.0       4175.1    331.4
Shell Scripts (1 concurrent)                     42.4       5677.1   1338.9
Shell Scripts (8 concurrent)                      6.0        770.9   1284.8
System Call Overhead                          15000.0    1860702.6   1240.5
                                                                   ========
System Benchmarks Index Score                                         769.6
------------------------------------------------------------------------

この無線設備は、電波法に定める技術基準への適合が確認されておらず、法に定める特別な条件の下でのみ使用が認められています。この条件に違反して無線設備を使用することは、法に定める罰則その他の措置の対象となります。

Copyright © ITmedia, Inc. All Rights Reserved.