Die folgenden Benchmarks wurden auf einem frisch installiertem Armbian 5.75 (Debian Stretch) ausgeführt. Es laufen daher keine Dienste wie Web- oder Datenbankserver im Hintergrund. Als Kernel kommt Linux Mainline 4.20.7 zum Einsatz.
CPU Performance
Als CPU-Benchmark nutze ich sysbench Version 0.4.12 aus den Debian-Paketquellen. Der Prozessor des BPI-M2 Berry ist ein ARM Cortex-A7 Quad-Core mit 1,5 GHz.
Single-Core Performance
sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 540.1434s
total number of events: 10000
total time taken by event execution: 540.1337
per-request statistics:
min: 53.99ms
avg: 54.01ms
max: 59.12ms
approx. 95 percentile: 54.03ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 540.1337/0.00
Multi-Core Performance (alle 4 Kerne)
sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 4
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 135.6720s
total number of events: 10000
total time taken by event execution: 542.5683
per-request statistics:
min: 53.99ms
avg: 54.26ms
max: 95.83ms
approx. 95 percentile: 54.22ms
Threads fairness:
events (avg/stddev): 2500.0000/13.40
execution time (avg/stddev): 135.6421/0.02
Netzwerk Performance
Der BPI-M2 Berry besitzt einen Gigabit-Ethernet Port. Ich verwende zum Testen iperf3. Der Berry ist über einen Switch mit meinem PC verbunden, der die iperf3-Serverrolle übernimmt.
iperf3 TX (Senderichtung)
iperf3 -c 192.168.178.10
Connecting to host 192.168.178.10, port 5201 [ 4] local 192.168.178.14 port 33090 connected to 192.168.178.10 port 5201 [ ID] Interval Transfer Bandwidth Retr Cwnd [ 4] 0.00-1.00 sec 99.4 MBytes 832 Mbits/sec 0 375 KBytes [ 4] 1.00-2.00 sec 98.8 MBytes 828 Mbits/sec 0 375 KBytes [ 4] 2.00-3.00 sec 98.7 MBytes 828 Mbits/sec 0 375 KBytes [ 4] 3.00-4.00 sec 98.6 MBytes 827 Mbits/sec 0 375 KBytes [ 4] 4.00-5.00 sec 98.6 MBytes 827 Mbits/sec 0 392 KBytes [ 4] 5.00-6.00 sec 98.4 MBytes 827 Mbits/sec 0 392 KBytes [ 4] 6.00-7.00 sec 98.4 MBytes 826 Mbits/sec 0 392 KBytes [ 4] 7.00-8.00 sec 98.6 MBytes 826 Mbits/sec 0 392 KBytes [ 4] 8.00-9.00 sec 98.6 MBytes 826 Mbits/sec 0 392 KBytes [ 4] 9.00-10.00 sec 98.2 MBytes 828 Mbits/sec 0 392 KBytes - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bandwidth Retr [ 4] 0.00-10.00 sec 986 MBytes 827 Mbits/sec 0 sender [ 4] 0.00-10.00 sec 985 MBytes 826 Mbits/sec receiver iperf Done.
iperf3 RX (Empfangsrichtung)
iperf3 -c 192.168.178.10 -R
Connecting to host 192.168.178.10, port 5201 Reverse mode, remote host 192.168.178.10 is sending [ 4] local 192.168.178.14 port 33100 connected to 192.168.178.10 port 5201 [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 109 MBytes 918 Mbits/sec [ 4] 1.00-2.00 sec 109 MBytes 919 Mbits/sec [ 4] 2.00-3.00 sec 110 MBytes 919 Mbits/sec [ 4] 3.00-4.00 sec 110 MBytes 919 Mbits/sec [ 4] 4.00-5.00 sec 110 MBytes 919 Mbits/sec [ 4] 5.00-6.00 sec 110 MBytes 919 Mbits/sec [ 4] 6.00-7.00 sec 109 MBytes 919 Mbits/sec [ 4] 7.00-8.00 sec 110 MBytes 919 Mbits/sec [ 4] 8.00-9.00 sec 110 MBytes 919 Mbits/sec [ 4] 9.00-10.00 sec 110 MBytes 919 Mbits/sec - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bandwidth Retr [ 4] 0.00-10.00 sec 1.07 GBytes 921 Mbits/sec 0 sender [ 4] 0.00-10.00 sec 1.07 GBytes 919 Mbits/sec receiver iperf Done.
SATA Performance
Zum Testen der Geschwindigkeit des SATA-Ports benutze ich eine ziemlich alte (2011) Kingston SSDNow V100 mit 64 GB (SATA II). Daher bitte die Messwerte mit Vorsicht genießen! Das Dateisystem habe ich frisch mit mkfs.ext4 formatiert. Da ich nicht die SSD, sondern die Schnittstelle testen will, messe ich nur die Geschwindingkeit beim sequentiellen Lesen.
Getestet wird mit 32 GB an Daten, um Caching zu vermeiden:
sysbench --test=fileio --file-total-size=32G prepare
sysbench --test=fileio --file-total-size=32G --file-test-mode=seqrd run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Extra file open flags: 0
128 files, 256Mb each
32Gb total file size
Block size 16Kb
Periodic FSYNC enabled, calling fsync() each 100 requests.
Calling fsync() at the end of test, Enabled.
Using synchronous I/O mode
Doing sequential read test
Threads started!
Done.
Operations performed: 2097152 Read, 0 Write, 0 Other = 2097152 Total
Read 32Gb Written 0b Total transferred 32Gb (132.56Mb/sec)
8483.82 Requests/sec executed
Test execution summary:
total time: 247.1944s
total number of events: 2097152
total time taken by event execution: 243.5119
per-request statistics:
min: 0.03ms
avg: 0.12ms
max: 11.79ms
approx. 95 percentile: 0.69ms
Threads fairness:
events (avg/stddev): 2097152.0000/0.00
execution time (avg/stddev): 243.5119/0.00
pappnaas
sehr hilfreich und nachvollziehbar.