|
|
|
Topic Category
|
|
|
|
Link Category
|
|
|
|
最新：04/16 19:55
|
|
|
|
最新：07/28 16:47
|
|
|
|
|
|
|
|
|
|
|
|
Raspberry Pi 2 Model B（４）
UNIX Benchmarkをやってみた
今度は4コア同時に高負荷をかけるべくベンチマーク UnixBench をやってみました。ベンチマークの結果は以下のとおり。
1コアの結果
Dhrystone 2 using register variables 3280874.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 552.4 MWIPS (9.9 s, 7 samples)
Execl Throughput 373.7 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 82969.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 23313.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 222100.5 KBps (30.0 s, 2 samples)
Pipe Throughput 192258.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching 35201.0 lps (10.0 s, 7 samples)
Process Creation 1472.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1224.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 351.9 lpm (60.1 s, 2 samples)
System Call Overhead 457212.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 3280874.2 281.1
Double-Precision Whetstone 55.0 552.4 100.4
Execl Throughput 43.0 373.7 86.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 82969.1 209.5
File Copy 256 bufsize 500 maxblocks 1655.0 23313.2 140.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 222100.5 382.9
Pipe Throughput 12440.0 192258.7 154.5
Pipe-based Context Switching 4000.0 35201.0 88.0
Process Creation 126.0 1472.1 116.8
Shell Scripts (1 concurrent) 42.4 1224.8 288.9
Shell Scripts (8 concurrent) 6.0 351.9 586.5
System Call Overhead 15000.0 457212.6 304.8
========
System Benchmarks Index Score 190.4
|
4コアの結果
Dhrystone 2 using register variables 13109791.0 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2210.6 MWIPS (9.9 s, 7 samples)
Execl Throughput 1554.4 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 123717.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 35832.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 351324.3 KBps (30.0 s, 2 samples)
Pipe Throughput 768381.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 135723.5 lps (10.0 s, 7 samples)
Process Creation 3257.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2654.5 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 373.3 lpm (60.3 s, 2 samples)
System Call Overhead 1761651.0 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 13109791.0 1123.4
Double-Precision Whetstone 55.0 2210.6 401.9
Execl Throughput 43.0 1554.4 361.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 123717.7 312.4
File Copy 256 bufsize 500 maxblocks 1655.0 35832.3 216.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 351324.3 605.7
Pipe Throughput 12440.0 768381.9 617.7
Pipe-based Context Switching 4000.0 135723.5 339.3
Process Creation 126.0 3257.2 258.5
Shell Scripts (1 concurrent) 42.4 2654.5 626.1
Shell Scripts (8 concurrent) 6.0 373.3 622.2
System Call Overhead 15000.0 1761651.0 1174.4
========
System Benchmarks Index Score 484.1
|
ベンチマーク実行中のCPUロード率は各コアとも最大70〜80%程度で、CPUの温度は62度まで上昇しました。しかし動作が不安定になることは無く、ベンチマークも正常に終了しました。
スコアを見ると、1コアでもB+の1.5倍程度高く、4コアでは4倍以上の値になっているようです。
|
|
|
2015 calendar
最新：08/15 17:19
|
|
