Premium Instances がリリースされました
Akamai's cloud computing servicesではリージョンを拡大しており、2023年7月12日に3リージョン(パリ、ワシントンD.C.、シカゴ)が開設されました。
同四半期中には、シアトルとチェンナイ(インド)を開設する予定です。
今回開設された3つのリージョンでは、新しいPremium Instancesタイプを利用することができます。
Akamai Cloud Managerからインスタンスを起動する際にPremiumのタブが選択できるようになっています。
価格やインスタンスのスペックについての情報はこちらになります。
Premium Instances とは
Akamai's cloud computing servicesでは各インスタンスのCPUにAMDのEYPC 7000シリーズが採用されています。
専有CPU型のDedicated InstancesでもこのシリーズのCPUが使われていますが、すべてに同一のCPU世代が採用されているわけではなく、7000シリーズ内のCPUが混在して利用されています。
これらのCPUはインスタンス起動時に空き状況を考慮して、最適なCPUが割り当てられるしくみになっています。
新たに用意されたPremium Instancesでは、利用可能なAMDのEYPC 7000シリーズのCPUの中で最新のCPUが割り当てられるよう設計されており、この他のインスタンスの仕様は既存のDedicated Instanceと同じです。
執筆時点では、Premium InstancesはAMD EPYC Milan以降のプロセッサ世代を最低保証しています。
これにより、高い処理性能を必要とするワークロードに適したCPUを常に選択できるようになりました。
例えば、1日中CPU利用率が100%になる動画のエンコーディングなどのユースケースに適しています。
その他インスタンスタイプについてはこちらを参照してください。
Premium Instances と Dedicated Instances を比較してみた
今回は一番性能が良いインスタンスを利用して、実際に性能が異なるのかをテストしてみました。
テストでは、UnixBench、fio、iperf3を利用して、CPU、ディスク、ネットワークのパフォーマンスを測定しています。
各リージョンのプロセッサは執筆者が起動した際に割り当てられたものであり、確実に該当のプロセッサが割当することを保証するものではありません。
また、本ベンチマーク結果はあくまでも参考値であり、結果を保証するものではありません。
測定に使ったコマンドと結果についてはこのページの最後に記載していますので、ご自身でも確認していただけます。
リージョン | インスタンスタイプ | CPUs | Processor | UnixBench 1 Parallel Score | UnixBentch 64 Parallel Score | fio Read | fio Write | iperf3 receiver Bitrate |
---|---|---|---|---|---|---|---|---|
Chicago | Premium 512 GB | 64 | AMD EPYC 7713 64-Core Processor | 1897.3 | 36636.6 | 611MiB/s | 610MiB/s | 11.7 Gbits/sec |
Sydney | Dedicated 512 GB | 64 | AMD EPYC 7642 48-Core Processor | 1722.8 | 35800.7 | 613MiB/s | 613MiB/s | 10.9 Gbits/sec |
AMD EPYC 7642 48-Core Processor場合には、1つのCPUで最大48コア利用できます。
Dedicated 512GBのインスタンスでは64コア必要となるため、ホストには2つのCPUが搭載されております。
上記結果から、CPUの世代が異なるため多少の性能差はあるももの、大きな性能差は出ませんでした。
そのため、高いパフォーマンスをどのインスタンスでも均一して維持したいユースケースにおいて、 Premium Instancesが活用できるかと思われます。
各リージョンの機能確認
今回、新たに追加されたPremium Plansは一部のリージョンでのみ提供しているサービスとなります。
各リージョンの機能を確認するにはドキュメントも参考になりますが、以下のようにLinode CLIを活用して確認することもできます。
各リージョンの機能は執筆時点の情報であり、最近情報は都度確認してください。
curl -s "https://gist.githubusercontent.com/isss802/ba5807f5f0694468004599aa27f5a771/raw/6cbc7a800d2e68f02f8341274b601367c307d95b/list-regions.py" > list-regions.py && linode-cli regions ls --json | jq '[.[] | del(.resolvers) | del(.status)]' | python3 list-regions.py && rm list-regions.py
+--------------+----------------+---------+---------+-------------+---------------+---------------+---------------+----------------+------------+----------------+-------+-------------------+
| ID | Label | Country | Linodes | GPU Linodes | Premium Plans | NodeBalancers | Block Storage | Object Storage | Kubernetes | Cloud Firewall | Vlans | Managed Databases |
+--------------+----------------+---------+---------+-------------+---------------+---------------+---------------+----------------+------------+----------------+-------+-------------------+
| ap-west | Mumbai, IN | in | ✔ | ✔ | - | ✔ | ✔ | - | ✔ | ✔ | ✔ | ✔ |
| ca-central | Toronto, CA | ca | ✔ | - | - | ✔ | ✔ | - | ✔ | ✔ | ✔ | ✔ |
| ap-southeast | Sydney, AU | au | ✔ | - | - | ✔ | ✔ | - | ✔ | ✔ | ✔ | ✔ |
| us-iad | Washington, DC | us | ✔ | - | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| us-ord | Chicago, IL | us | ✔ | - | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| fr-par | Paris, FR | fr | ✔ | - | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| us-central | Dallas, TX | us | ✔ | - | - | ✔ | ✔ | - | ✔ | ✔ | - | ✔ |
| us-west | Fremont, CA | us | ✔ | - | - | ✔ | ✔ | - | ✔ | ✔ | - | ✔ |
| us-southeast | Atlanta, GA | us | ✔ | ✔ | - | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| us-east | Newark, NJ | us | ✔ | ✔ | - | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| eu-west | London, UK | uk | ✔ | - | - | ✔ | ✔ | - | ✔ | ✔ | ✔ | ✔ |
| ap-south | Singapore, SG | sg | ✔ | ✔ | - | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| eu-central | Frankfurt, DE | de | ✔ | ✔ | - | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| ap-northeast | Tokyo, JP | jp | ✔ | - | - | ✔ | ✔ | - | ✔ | ✔ | - | ✔ |
+--------------+----------------+---------+---------+-------------+---------------+---------------+---------------+----------------+------------+----------------+-------+-------------------+
まとめ
Akamai's cloud computing servicesでは、リージョンの拡張とともに新機能も続々とリリース予定です。
ぜひ、新しい機能を活用して、様々なワークロードで活用してみてください。
関連記事
アカマイ・テクノロジーズ合同会社はQiitaでAkamai's cloud computing services関連など開発者向けの記事を掲載しております。
Premium 512 GB の測定結果詳細
root@localhost:~# sudo lshw -class processor
*-cpu
description: CPU
product: AMD EPYC 7713 64-Core Processor
vendor: Advanced Micro Devices [AMD]
physical id: 400
bus info: cpu@0
version: 25.1.1
slot: CPU 0
size: 2GHz
capacity: 2GHz
width: 64 bits
capabilities: fpu fpu_exception wp vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp x86-64 rep_good nopl cpuid extd_apicid tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw perfctr_core ssbd ibrs ibpb stibp vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves clzero xsaveerptr wbnoinvd arat umip pku ospke vaes vpclmulqdq rdpid arch_capabilities
configuration: cores=64 enabledcores=64 microcode=167776627 threads=1
root@localhost:~/byte-unixbench/UnixBench# ./Run -i 5
Benchmark Run: Thu Jul 13 2023 15:55:38 - 16:12:25 [35/1336]
64 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 42640262.3 lps (10.0 s, 4 samples)
Double-Precision Whetstone 7511.8 MWIPS (10.0 s, 4 samples)
Execl Throughput 3972.4 lps (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1310125.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 347058.7 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 3919541.1 KBps (30.0 s, 2 samples)
Pipe Throughput 2039441.6 lps (10.0 s, 4 samples)
Pipe-based Context Switching 83606.1 lps (10.0 s, 4 samples)
Process Creation 8424.9 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 10557.1 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 7674.1 lpm (60.0 s, 2 samples)
System Call Overhead 2068987.0 lps (10.0 s, 4 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 42640262.3 3653.8
Double-Precision Whetstone 55.0 7511.8 1365.8
Execl Throughput 43.0 3972.4 923.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 1310125.0 3308.4
File Copy 256 bufsize 500 maxblocks 1655.0 347058.7 2097.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 3919541.1 6757.8
Pipe Throughput 12440.0 2039441.6 1639.4
Pipe-based Context Switching 4000.0 83606.1 209.0
Process Creation 126.0 8424.9 668.6
Shell Scripts (1 concurrent) 42.4 10557.1 2489.9
Shell Scripts (8 concurrent) 6.0 7674.1 12790.2
System Call Overhead 15000.0 2068987.0 1379.3
========
System Benchmarks Index Score 1897.3
------------------------------------------------------------------------
Benchmark Run: Thu Jul 13 2023 16:12:25 - 16:29:20
64 CPUs in system; running 64 parallel copies of tests
Dhrystone 2 using register variables 1784050324.6 lps (10.0 s, 4 samples)
Double-Precision Whetstone 411946.2 MWIPS (9.8 s, 4 samples)
Execl Throughput 20641.4 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 29849573.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 15335643.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 21530007.2 KBps (30.0 s, 2 samples)
Pipe Throughput 89720706.0 lps (10.0 s, 4 samples)
Pipe-based Context Switching 10067861.9 lps (10.0 s, 4 samples)
Process Creation 63975.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 113003.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 15174.1 lpm (60.1 s, 2 samples)
System Call Overhead 99071822.4 lps (10.0 s, 4 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 1784050324.6 152874.9
Double-Precision Whetstone 55.0 411946.2 74899.3
Execl Throughput 43.0 20641.4 4800.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 29849573.1 75377.7
File Copy 256 bufsize 500 maxblocks 1655.0 15335643.2 92662.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 21530007.2 37120.7
Pipe Throughput 12440.0 89720706.0 72122.8
Pipe-based Context Switching 4000.0 10067861.9 25169.7
Process Creation 126.0 63975.8 5077.4
Shell Scripts (1 concurrent) 42.4 113003.4 26651.8
Shell Scripts (8 concurrent) 6.0 15174.1 25290.1
System Call Overhead 15000.0 99071822.4 66047.9
========
System Benchmarks Index Score 36636.6
root@localhost:~/byte-unixbench/UnixBench# fio --name=tempfile --size=10G --runtime=2m --rw=readwrite --loops=10
tempfile: (g=0): rw=rw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=psync, iodepth=1
fio-3.28
Starting 1 process
Jobs: 1 (f=1): [M(1)][100.0%][eta 00m:00s]
tempfile: (groupid=0, jobs=1): err= 0: pid=2942021: Thu Jul 13 16:56:13 2023
read: IOPS=156k, BW=611MiB/s (640MB/s)(50.0GiB/83863msec)
clat (nsec): min=330, max=23418k, avg=3422.49, stdev=32605.18
lat (nsec): min=360, max=23418k, avg=3459.92, stdev=32605.22
clat percentiles (nsec):
| 1.00th=[ 422], 5.00th=[ 490], 10.00th=[ 524], 20.00th=[ 532],
| 30.00th=[ 532], 40.00th=[ 540], 50.00th=[ 548], 60.00th=[ 548],
| 70.00th=[ 564], 80.00th=[ 572], 90.00th=[ 628], 95.00th=[ 732],
| 99.00th=[ 57088], 99.50th=[292864], 99.90th=[362496], 99.95th=[370688],
| 99.99th=[411648]
bw ( KiB/s): min=67832, max=984768, per=100.00%, avg=743670.41, stdev=131470.16, samples=141
iops : min=16958, max=246194, avg=185917.73, stdev=32867.67, samples=141
write: IOPS=156k, BW=610MiB/s (640MB/s)(50.0GiB/83863msec); 0 zone resets
clat (nsec): min=760, max=2184.1k, avg=1207.30, stdev=4409.44
lat (nsec): min=800, max=2184.1k, avg=1258.36, stdev=4410.64
clat percentiles (nsec):
| 1.00th=[ 940], 5.00th=[ 964], 10.00th=[ 980], 20.00th=[ 1020],
| 30.00th=[ 1048], 40.00th=[ 1048], 50.00th=[ 1064], 60.00th=[ 1096],
| 70.00th=[ 1112], 80.00th=[ 1128], 90.00th=[ 1288], 95.00th=[ 1624],
| 99.00th=[ 2416], 99.50th=[ 2928], 99.90th=[ 8896], 99.95th=[ 11328],
| 99.99th=[296960]
bw ( KiB/s): min=68672, max=985608, per=100.00%, avg=743134.64, stdev=131779.35, samples=141
iops : min=17168, max=246402, avg=185783.69, stdev=32944.90, samples=141
lat (nsec) : 500=2.55%, 750=45.10%, 1000=8.02%
lat (usec) : 2=42.18%, 4=1.16%, 10=0.20%, 20=0.03%, 50=0.11%
lat (usec) : 100=0.22%, 250=0.12%, 500=0.32%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.01%, 4=0.01%, 10=0.01%, 20=0.01%, 50=0.01%
cpu : usr=12.41%, sys=49.92%, ctx=170244, majf=0, minf=22
IO depths : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=13111888,13102512,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: bw=611MiB/s (640MB/s), 611MiB/s-611MiB/s (640MB/s-640MB/s), io=50.0GiB (53.7GB), run=83863-83863msec
WRITE: bw=610MiB/s (640MB/s), 610MiB/s-610MiB/s (640MB/s-640MB/s), io=50.0GiB (53.7GB), run=83863-83863msec
Disk stats (read/write):
sda: ios=197555/61247, merge=0/35, ticks=87263/147089, in_queue=234358, util=88.91%
root@localhost:~# iperf3 -c 192.168.130.38
Connecting to host 192.168.130.38, port 5201
[ 5] local 192.168.129.93 port 47062 connected to 192.168.130.38 port 5201
[ ID] Interval Transfer Bitrate Retr Cwnd
[ 5] 0.00-1.00 sec 1.40 GBytes 12.0 Gbits/sec 0 3.14 MBytes
[ 5] 1.00-2.00 sec 1.37 GBytes 11.7 Gbits/sec 0 3.14 MBytes
[ 5] 2.00-3.00 sec 1.37 GBytes 11.7 Gbits/sec 0 3.14 MBytes
[ 5] 3.00-4.00 sec 1.37 GBytes 11.8 Gbits/sec 0 3.14 MBytes
[ 5] 4.00-5.00 sec 1.37 GBytes 11.8 Gbits/sec 0 3.14 MBytes
[ 5] 5.00-6.00 sec 1.40 GBytes 12.0 Gbits/sec 0 3.14 MBytes
[ 5] 6.00-7.00 sec 1.39 GBytes 12.0 Gbits/sec 0 3.14 MBytes
[ 5] 7.00-8.00 sec 1.38 GBytes 11.9 Gbits/sec 0 3.14 MBytes
[ 5] 8.00-9.00 sec 1.34 GBytes 11.5 Gbits/sec 0 3.14 MBytes
[ 5] 9.00-10.00 sec 1.34 GBytes 11.5 Gbits/sec 0 3.14 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bitrate Retr
[ 5] 0.00-10.00 sec 13.7 GBytes 11.8 Gbits/sec 0 sender
[ 5] 0.00-10.04 sec 13.7 GBytes 11.7 Gbits/sec receiver
Dedicated 512 GBの測定結果詳細
root@localhost:~# sudo lshw -class processor
*-cpu
description: CPU
product: AMD EPYC 7642 48-Core Processor
vendor: Advanced Micro Devices [AMD]
physical id: 400
bus info: cpu@0
version: 23.49.0
slot: CPU 0
size: 2GHz
capacity: 2GHz
width: 64 bits
capabilities: fpu fpu_exception wp vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp x86-64 rep_good nopl cpuid extd_apicid tsc_known_freq pni pclmulqdq ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw perfctr_core ssbd ibrs ibpb stibp vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves clzero xsaveerptr wbnoinvd arat umip rdpid arch_capabilities
configuration: cores=64 enabledcores=64 microcode=137367637 threads=1
root@localhost:~/byte-unixbench/UnixBench# ./Run -i 5
Benchmark Run: Thu Jul 13 2023 15:55:47 - 16:12:34
64 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 39956317.8 lps (10.0 s, 4 samples)
Double-Precision Whetstone 7167.5 MWIPS (10.0 s, 4 samples)
Execl Throughput 4025.6 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1120617.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 300097.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2864270.0 KBps (30.0 s, 2 samples)
Pipe Throughput 1654884.0 lps (10.0 s, 4 samples)
Pipe-based Context Switching 75669.1 lps (10.0 s, 4 samples)
Process Creation 8067.3 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 10547.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 8184.9 lpm (60.0 s, 2 samples)
System Call Overhead 1771302.2 lps (10.0 s, 4 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 39956317.8 3423.8
Double-Precision Whetstone 55.0 7167.5 1303.2
Execl Throughput 43.0 4025.6 936.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 1120617.3 2829.8
File Copy 256 bufsize 500 maxblocks 1655.0 300097.0 1813.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 2864270.0 4938.4
Pipe Throughput 12440.0 1654884.0 1330.3
Pipe-based Context Switching 4000.0 75669.1 189.2
Process Creation 126.0 8067.3 640.3
Shell Scripts (1 concurrent) 42.4 10547.5 2487.6
Shell Scripts (8 concurrent) 6.0 8184.9 13641.5
System Call Overhead 15000.0 1771302.2 1180.9
========
System Benchmarks Index Score 1722.8
------------------------------------------------------------------------
Benchmark Run: Thu Jul 13 2023 16:12:34 - 16:29:28
64 CPUs in system; running 64 parallel copies of tests
Dhrystone 2 using register variables 1743568460.1 lps (10.0 s, 4 samples)
Double-Precision Whetstone 407292.5 MWIPS (9.9 s, 4 samples)
Execl Throughput 24130.2 lps (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 41334086.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 13940281.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 28204710.7 KBps (30.0 s, 2 samples)
Pipe Throughput 80955959.6 lps (10.0 s, 4 samples)
Pipe-based Context Switching 3905313.7 lps (10.0 s, 4 samples)
Process Creation 63534.3 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 130929.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 17628.1 lpm (60.1 s, 2 samples)
System Call Overhead 86182676.0 lps (10.0 s, 4 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 1743568460.1 149406.0
Double-Precision Whetstone 55.0 407292.5 74053.2
Execl Throughput 43.0 24130.2 5611.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 41334086.5 104379.0
File Copy 256 bufsize 500 maxblocks 1655.0 13940281.4 84231.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 28204710.7 48628.8
Pipe Throughput 12440.0 80955959.6 65077.1
Pipe-based Context Switching 4000.0 3905313.7 9763.3
Process Creation 126.0 63534.3 5042.4
Shell Scripts (1 concurrent) 42.4 130929.8 30879.7
Shell Scripts (8 concurrent) 6.0 17628.1 29380.1
System Call Overhead 15000.0 86182676.0 57455.1
========
System Benchmarks Index Score 35800.7
root@localhost:~/byte-unixbench/UnixBench# fio --name=tempfile --size=10G --runtime=2m --rw=readwrite --loops=10
tempfile: (g=0): rw=rw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=psync, iodepth=1
fio-3.28
Starting 1 process
Jobs: 1 (f=1): [M(1)][100.0%][eta 00m:00s]
tempfile: (groupid=0, jobs=1): err= 0: pid=1561268: Thu Jul 13 17:03:43 2023
read: IOPS=157k, BW=613MiB/s (643MB/s)(50.0GiB/83561msec)
clat (nsec): min=460, max=53785k, avg=2562.10, stdev=66093.16
lat (nsec): min=500, max=53785k, avg=2606.85, stdev=66093.28
clat percentiles (nsec):
| 1.00th=[ 490], 5.00th=[ 604], 10.00th=[ 612], 20.00th=[ 620],
| 30.00th=[ 628], 40.00th=[ 644], 50.00th=[ 652], 60.00th=[ 660],
| 70.00th=[ 684], 80.00th=[ 700], 90.00th=[ 740], 95.00th=[ 804],
| 99.00th=[ 50432], 99.50th=[ 51456], 99.90th=[444416], 99.95th=[610304],
| 99.99th=[880640]
bw ( KiB/s): min=24576, max=1197416, per=100.00%, avg=812782.36, stdev=367312.95, samples=129
iops : min= 6144, max=299354, avg=203195.64, stdev=91828.28, samples=129
write: IOPS=157k, BW=613MiB/s (642MB/s)(50.0GiB/83561msec); 0 zone resets
clat (nsec): min=1020, max=3560.3k, avg=1418.79, stdev=4456.17
lat (nsec): min=1080, max=3560.4k, avg=1479.94, stdev=4457.61
clat percentiles (nsec):
| 1.00th=[ 1176], 5.00th=[ 1208], 10.00th=[ 1224], 20.00th=[ 1240],
| 30.00th=[ 1256], 40.00th=[ 1272], 50.00th=[ 1288], 60.00th=[ 1304],
| 70.00th=[ 1304], 80.00th=[ 1336], 90.00th=[ 1576], 95.00th=[ 1768],
| 99.00th=[ 2768], 99.50th=[ 7072], 99.90th=[ 8512], 99.95th=[10688],
| 99.99th=[26752]
bw ( KiB/s): min=24984, max=1197736, per=100.00%, avg=812196.64, stdev=367081.37, samples=129
iops : min= 6246, max=299434, avg=203049.14, stdev=91770.33, samples=129
lat (nsec) : 500=0.73%, 750=44.67%, 1000=3.48%
lat (usec) : 2=48.19%, 4=1.63%, 10=0.51%, 20=0.03%, 50=0.21%
lat (usec) : 100=0.44%, 250=0.02%, 500=0.04%, 750=0.03%, 1000=0.01%
lat (msec) : 2=0.01%, 4=0.01%, 10=0.01%, 20=0.01%, 50=0.01%
lat (msec) : 100=0.01%
cpu : usr=15.59%, sys=51.27%, ctx=32001, majf=0, minf=21
IO depths : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=13111888,13102512,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: bw=613MiB/s (643MB/s), 613MiB/s-613MiB/s (643MB/s-643MB/s), io=50.0GiB (53.7GB), run=83561-83561msec
WRITE: bw=613MiB/s (642MB/s), 613MiB/s-613MiB/s (642MB/s-642MB/s), io=50.0GiB (53.7GB), run=83561-83561msec
Disk stats (read/write):
sda: ios=197545/41678, merge=0/22, ticks=58429/347904, in_queue=406336, util=90.97%
root@localhost:~# iperf3 -c 192.168.136.21
Connecting to host 192.168.136.21, port 5201
[ 5] local 192.168.131.15 port 42526 connected to 192.168.136.21 port 5201
[ ID] Interval Transfer Bitrate Retr Cwnd
[ 5] 0.00-1.00 sec 1.33 GBytes 11.4 Gbits/sec 0 3.13 MBytes
[ 5] 1.00-2.00 sec 1.30 GBytes 11.1 Gbits/sec 0 3.13 MBytes
[ 5] 2.00-3.00 sec 1.29 GBytes 11.1 Gbits/sec 0 3.13 MBytes
[ 5] 3.00-4.00 sec 1.29 GBytes 11.0 Gbits/sec 0 3.13 MBytes
[ 5] 4.00-5.00 sec 1.28 GBytes 11.0 Gbits/sec 0 3.13 MBytes
[ 5] 5.00-6.00 sec 1.27 GBytes 10.9 Gbits/sec 0 3.13 MBytes
[ 5] 6.00-7.00 sec 1.26 GBytes 10.9 Gbits/sec 0 3.13 MBytes
[ 5] 7.00-8.00 sec 1.26 GBytes 10.9 Gbits/sec 0 3.13 MBytes
[ 5] 8.00-9.00 sec 1.26 GBytes 10.8 Gbits/sec 0 3.13 MBytes
[ 5] 9.00-10.00 sec 1.26 GBytes 10.8 Gbits/sec 0 3.13 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bitrate Retr
[ 5] 0.00-10.00 sec 12.8 GBytes 11.0 Gbits/sec 0 sender
[ 5] 0.00-10.04 sec 12.8 GBytes 10.9 Gbits/sec receiver