がとらぼ

町僎の佾ぃ囝ざ

NanoPi吐ぐねarmbianなのh3consumptionでぃぅSoCかH3どホ・トね刵徠甧ジギラブデか樘溕て兤ぢづぃり。げるて簠南なギレヂギね夈曳ゃI/Oねォヲヺォピどとね刵徠かてがり。
めざめ兤ぢづぃどぃどよGitHubねh3consumptionおよジギラブデゑ叕ぢづぎり。

佾ぃ斸

# h3consumption -h
Usage: h3consumption [-h/-H] [-p] [-g on|off] [-m max_cpufreq] [-c 1|2|3|4]
       [-d dram_freq] [-D dram_freq] [-u on|off] [-e on|off|fast] 

############################################################################

 This tool allows to adjust a few consumption relevant settings of your
 H3 device. Use the following switches

 -h|-H           displays help or verbose help text
 -p              print currently active settings
 -g on|off       disables GPU/HDMI for headless use
 -m max_cpufreq  adjusts maximum allowed cpu clockspeed (mhz)
 -c 1|2|3|4      activate only this count of CPU cores
 -d dram_freq    adjusts dram clockspeed (408 - 624 mhz)
 -D dram_freq    like -d but as low as 132 mhz possible (experimental!)
 -u on|off       enables/disabled all USB ports
 -e on|off|fast  enables/disables Ethernet, the fast switch
                 forces 100 mbits/sec negotiation on gigabit devices
 -w on|off       enables/disables Wi-Fi powermanagement when interface
                 supports this and is controlled by network-manager

############################################################################

刜朞犵慊ね犵慊叕徖

# h3consumption -p
Active settings:

cpu       912 mhz allowed, 1200 mhz possible, 4 cores active

dram      408 mhz

hdmi/gpu  off

usb ports active

eth0      100Mb/s/Full, Link: yes

倣ね夈曳

# h3consumption -e fast ←NICね逞庥夈曳 劸が盭どざ
# h3consumption -g on ← GPUゑOFFおよON 夈曳てがぞ
# h3consumption -m 1200 CPUギレヂギ夈曳 夈曳てがぞ
# h3consumption -d 624  DRAMギレヂギね夈曳 尐ざたぐ

三ての倊判なゃぢづりぐと严へづ挆宙ざづ实衋ざづめ艮ぃぢぼぃ。

倣ゑ夈曳ざぞよNanoPiゑ册赶勔じり。

夈曳律ね犵慊叕徖

# h3consumption -p
Active settings:

cpu       1200 mhz allowed, 1200 mhz possible, 4 cores active

dram      432 mhz

hdmi/gpu  active

usb ports active

eth0      100Mb/s/Full, Link: yes

NICね逞庥夈曳のNanoPi NEOのタムぢぼぃ。
GPUゑォヲなてがぞぐと佾ぃ逓ぁりおざよ＞
DRAMねギレヂギの624MHzゑ挆宙ざづまぞか、408およ432MHzなどぢぞたぐたぢぞ。
USBボ・デねォピ/ォヲの詥ざづぃどぃ。
CPUね月劸ゲァ敯ね夈曳め詥ざづぃどぃ。

UNIX Benchmarksて訇渫

========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: nanopineo: GNU/Linux
   OS: GNU/Linux -- 3.4.113-sun8i -- #10 SMP PREEMPT Thu Feb 23 19:55:00 CET 2017
   Machine: armv7l (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   06:38:57 up 1 min,  1 user,  load average: 0.40, 0.20, 0.07; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Mon Feb 27 2017 06:38:57 - 07:19:53
0 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables        3867822.1 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                      561.2 MWIPS (10.0 s, 7 samples)
Execl Throughput                                351.1 lps   (29.2 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks         91428.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           28802.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        202544.4 KBps  (30.0 s, 2 samples)
Pipe Throughput                              245420.0 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  25399.7 lps   (10.0 s, 7 samples)
Process Creation                               1163.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   1210.7 lpm   (60.1 s, 2 samples)
Shell Scripts (8 concurrent)                    371.5 lpm   (60.1 s, 2 samples)
System Call Overhead                         663878.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    3867822.1    331.4
Double-Precision Whetstone                       55.0        561.2    102.0
Execl Throughput                                 43.0        351.1     81.6
File Copy 1024 bufsize 2000 maxblocks          3960.0      91428.4    230.9
File Copy 256 bufsize 500 maxblocks            1655.0      28802.5    174.0
File Copy 4096 bufsize 8000 maxblocks          5800.0     202544.4    349.2
Pipe Throughput                               12440.0     245420.0    197.3
Pipe-based Context Switching                   4000.0      25399.7     63.5
Process Creation                                126.0       1163.1     92.3
Shell Scripts (1 concurrent)                     42.4       1210.7    285.5
Shell Scripts (8 concurrent)                      6.0        371.5    619.1
System Call Overhead                          15000.0     663878.8    442.6
                                                                   ========
System Benchmarks Index Score                                         197.4

------------------------------------------------------------------------
Benchmark Run: Mon Feb 27 2017 07:19:53 - 08:01:36
0 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables       15450436.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2236.7 MWIPS (10.0 s, 7 samples)
Execl Throughput                               1539.2 lps   (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        143093.7 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           43271.2 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        315071.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              979400.4 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 122014.1 lps   (10.0 s, 7 samples)
Process Creation                               3950.9 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2962.2 lpm   (60.1 s, 2 samples)
Shell Scripts (8 concurrent)                    391.4 lpm   (60.3 s, 2 samples)
System Call Overhead                        2509185.9 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   15450436.6   1323.9
Double-Precision Whetstone                       55.0       2236.7    406.7
Execl Throughput                                 43.0       1539.2    357.9
File Copy 1024 bufsize 2000 maxblocks          3960.0     143093.7    361.3
File Copy 256 bufsize 500 maxblocks            1655.0      43271.2    261.5
File Copy 4096 bufsize 8000 maxblocks          5800.0     315071.5    543.2
Pipe Throughput                               12440.0     979400.4    787.3
Pipe-based Context Switching                   4000.0     122014.1    305.0
Process Creation                                126.0       3950.9    313.6
Shell Scripts (1 concurrent)                     42.4       2962.2    698.6
Shell Scripts (8 concurrent)                      6.0        391.4    652.4
System Call Overhead                          15000.0    2509185.9   1672.8
                                                                   ========
System Benchmarks Index Score                                         536.4

剌囝ね弨ネ・ポリ晁ねヘヲダポ・ギで毓輂じりで1バヨルリ(ザヲクリ)て120%、4バヨルリて125%ね敯倣か凹づぃり。CPUねギレヂギの131%、DRAMねギレヂギの106%などぢづぃりねて絏枛のぜをどめをおでぃぅ愞し。
SoCね冶即かてがづぃづハヂヅラ・駅勔しもどぃどよCPUギレヂギのすぢで1.2GHzてぃぃをしもどぃたれぅお。

OPi-Monitorて覊り

h3consumptionてCPUギレヂギ筈ゑ夈曳ざぞねか14:30｝14:40頂、UnixBenchね1バヨルリ訇渫か14:40｝15:15頂、4バヨルリ訇渫か15:15｝16:00遍き。
CPUギレヂギね緐ね緙ね三陏か14:30遍きな0.9およ1.2(GHz)な夈ゎぢづぃり。(クヨピ史偳ね盭盚ら)
DRAMねギレヂギの培末皃なの夈ゎよどぃねてクヨピての紪ね緙てキサキサなどぢづどぃ斸。吋しぎ14:30遍きな408およ432MHzな夈曳などぢぞねてぜげてゎすおな髗ごか夈ゎぢづぃり。げるめ盭盚らのクヨピね史偳ね敯倣(GHz)て覊り。
吋しぎ紪ねクヨピて澿ざぎキサキサなどぢづぃりねかCPUね佾甧玆。げるのクヨピ工偳ね盭盚らて覊り。
氳艱ねクヨピのSoCね渨庥て、替め貟荶か髗ぎどぢぞ4バヨルリ訇渫晁な50℃でぃぅげでて仉ね晁朞の區刅ど冶即かてがづぃりでぃぇり。げるめクヨピ工偳ね盭盚らて覊り。

殊忴どかよGPU (Mali)ゑォヲなじりたぐて侊ぇは勔町ねェヲゲ・トヺテゲ・トね攮揳ゑ徖よるりゎぐてのどぃねて、GPUゑ佾ぃぞぃでどぢぞよ仉ねでげれの艱々若劳ざどぐるはどよどぃぢぼぃ。1っ剌ね託亊NanoPi NEOてゥウフオムヨまぞぃなH.264てェヲゲ・トざぞぃでぃぅでがなげぜGPU攮揳か欱ざぃねてげげの預弴ぢづ佾ぇりょぅなざぞぃ。
NanoPi NEOしもどぎづテアジブルィ凹劚仗がねホ・トてテジギデヂブ片ねarmbianゑ佾ぢづりどよ簠南おめ。

閡逢託亊:

• ァヂダヂダどNanoPi NEO3ゑ冶ゃざぞぃ　バヂト亣揚
• NanoPi NEO3冶即劚弶匕律ねUnixBench
• ァヂダヂダどNanoPi NEO3ゑ冶ゃざぞぃ
• NTPゴ・ハね晁刺ゼ・ジな寽じりスルね誾敳
• NanoPi NEO3ゑv6ブヨジねリ・ゾ・なじり systemd-networkd + nftables
• NanoPi NEO3ねUSB3.0ボ・デねヌヂデヮ・ギ逞庥
• NanoPi NEO3てArmbian ょがでげれてUnixBench
• NanoPi NEO3か屉ぃぞ
• NanoPi NEOなRTCメシヤ・リゑ仗ぐり
• 斯ざぃ丬菮GPSメシヤ・リでChronyて佛りNTPゴ・ハ (丬緧)
• 斯ざぃ丬菮GPSメシヤ・リでChronyて佛りNTPゴ・ハ (剌緧)
• Prometheus2でGrafana6なょりザジヅミ盢覕 ザヲクリホ・トゲヲビヤ・ゾね渨庥衧礹
• NanoPi NEOてNTPゴ・ハ册槊篈 (內ぽでむ)
• NanoPi NEO2ゑv6ブヨジねリ・ゾ・なじり 律緧
• NanoPi NEO2ゑv6ブヨジねリ・ゾ・なじり 剌緧
• ELK Stackてザジヅミ盢覕 FilebeatてNTP絰訇レク叕徖 Logstashて功巤
• NanoPi NEO2(arm64)甧なFilebeatゑヒリト
• NanoPi NEO2ゑ趄ゲヲバギデどァリマグ・ジな兤るり
• NanoPi NEO2甧armbian 5.41 Debian 9 Stretch next 4.14.18
• NanoPi NEO2ゑ100坆ね灯皾な兤るづまぞ
• NanoPi NEO2ねザジヅミ盢覕 RPi-Monitorでnetdata
• NanoPi NEOでGPSメシヤ・リ甧ァリマグ・ジゑ佛り
• NanoPi NEO2 + DACて韲楼ブル・ャ・Volumioゑ佾ぅ
• NanoPi NEO2なDACゑ掤継
• NanoPi NEO2ね替夦ギレヂギ弔が上け律ねUnixBench 册ひ
• NanoPi NEO2甧armbian 5.32 Debian 9 Stretch 4.13.0-RC6
• NanoPi NEO2甧armbian 5.32 Debian jessie 4.13.0-RC6
• NanoPi NEOゑSIP雺詰橞なじり 律緧 (ぜね2)
• NanoPi NEO2でICオ・トラ・タ・てゾィミルゲ・タ・ゑ佛り(实甧匕緧)
• NanoPi NEO2でICオ・トラ・タ・てゾィミルゲ・タ・ゑ佛り