安価な仮想サーバーのテスト

多くのホスティング会社は安価な仮想サーバーを販売しており、最近ではさまざまな制限付きの広告料金 (たとえば、100 つのアカウントに対してそのような仮想サーバーを XNUMX つ注文できる) が大量に表示され始めており、その価格は場合によっては定価よりも安い場合もあります。 IP アドレスのコスト。 ちょっとしたテストを実施して、その結果を広く一般の人々と共有することが興味深くなりました。 仮想サーバーの一部は、テストを投稿する私のプロジェクトの XNUMX つのためにホスティング業者によって以前に提供されたものであり、さらに、一定数の VPS が XNUMX 台あたり XNUMX ルーブル以下で購入されました。

安価な仮想サーバーのテスト

仮想サーバーの特性は互いに異なり、ノードまたはチャネルの負荷、ノード上のクライアント数、時間に応じて、特定の時点で測定されたパフォーマンスは非常に相対的な値であることにすぐに注意してください。一日の気温や梅雨時の火星の平均気温など、なかなか面白い内容です。

オペレーティング システムとして Debian 9 を選択し、テストには unixbench、dd、fio、speedtest などのいくつかの人気のあるプログラムを使用します。 dd を使用して、仮想マシンごとに 64 つのテスト (1K ブロックと 4M ブロック) が実行されました。 fio ユーティリティは、ランダム読み取り、ランダム直接読み取り、ランダム書き込み、およびランダム直接書き込みモードで XNUMX 回実行されました。

結果の前に、料金と価格について簡単に説明します。

ホスター
ディスク容量 / CPUリソース / RAM
Цена

RUVDS
10 GB SSD
1x 2 GHz (インテル Xeon E5-2660v4)
0.5 GB (メモリ合計: 428980 KB)
30摩擦/月

timeweb
5 GB SSD
1x 2.4 GHz (vCPU)
0.5 GB (メモリ合計: 504192 KB)
45摩擦/月

イノベンティカサービス
5 GB HDD
1x 2 GHz (vCPU)
0.5 GB (メモリ合計: 504188 KB)
49摩擦/月

ホスティング-ロシア
10 GB SSD
1x 3.3 GHz (vCPU)
0.5 GB (メモリ合計: 504308 KB)
50摩擦/月

最初のバイト
7 GB SSD
1x 2.9 GHz (vCPU)
0.5 GB (メモリ合計: 504188 KB)
55摩擦/月

VDSina
5 GB NVMe SSD
NL - 1x 2.1 GHz (vCPU);
RU – 1x 2.2 GHz (vCPU)
RU、ニュージャージー州 – 0.5 GB (MemTotal: 504328 kB)
59摩擦/月

ゾムロ
5 GB SSD
1x 2.13 GHz (インテル Xeon L5630)
0.25 GB (メモリ合計: 262144 KB)
$0.99/月 (~63 ルーブル)

ヴィースプ
10 GB HDD
1x 2.67 GHz (インテル Xeon X5650)
0.5 GB (メモリ合計: 504328 KB)
65摩擦/月

PQ.ホスティング
15 GB SSD
RU – 1x 2.8 GHz (vCPU);
NL - 1x 2 GHz (vCPU);
MD - 1x 2.8 GHz (vCPU);
LV – 1x 2.8 GHz (vCPU)
RU - 1 GB (MemTotal: 995824 KB);
NL - 1 GB (MemTotal: 1020400 kB);
MD - 1 GB (MemTotal: 995820 kB);
LV - 1 GB (メモリ合計: 995816 kB)
1 ユーロ/月 (約 70 ルーブル)

グッドホスト
5 GB SSD
1x 4.2 GHz (vCPU)
0.4 GB (メモリ合計: 389632 KB)
79摩擦/月

管理VPS
10 GB SSD
1x 2.5 GHz (インテル Xeon E5-2680v3)
0.5 GB (メモリ合計: 524288 KB)
90摩擦/月

ファーストVDS
10 GB HDD
1x 1.6 GHz (インテル Xeon E5645)
0.5 GB (メモリ合計: 524288 KB)
90摩擦/月

VPSホスター
5 GB SSD
1x 2.26 GHz (vCPU)
0.5 GB (メモリ合計: 479612 KB)
99摩擦/月

MyVPS
10 GB HDD + SSD キャッシュ
1x 0.6 GHz (インテル Xeon E5-2670)
0.25 GB (メモリ合計: 262144 KB)
99摩擦/月

イホル
10 GB HDD
1x 2.4 GHz (インテル Xeon E5620)
0.5 GB (メモリ合計: 504280 KB)
100摩擦/月

ほとんどの仮想サーバーはロシアにありますが、例外もあります。

それでは、テストに移りましょう。文字の壁を読みたくない人のために、次のボタンを残しておきます。 結果 (グラフがたくさんあります)。

RUVDS の仮想サーバー:

安価な仮想サーバーのテスト

  • 装置はモスクワ近郊のコロレフの町にあります(このデータセンターを訪問したことについてはすでに書きました) );
  • 格安料金 (10 GB SSD / 1x 2 GHz CPU / 0.5 GB RAM / 100 Mbit/s で月額 30 ルーブル)
  • 現時点ではホスティング業者はまだハブラ効果に苦しんでおり、この料金は事前注文によってのみ利用可能です。
  • このような VPS はアカウントごとに XNUMX つまでに制限されています。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-8-amd64 -- #1 SMP Debian 4.9.130-2 (2018-10-27)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Xeon(R) CPU E5-2660 v4 @ 2.00GHz (3999.9 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   12:42:09 up 9 min,  1 user,  load average: 0.80, 0.33, 0.12; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Thu Aug 29 2019 12:42:09 - 13:10:36
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       22313053.1 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3909.2 MWIPS (9.9 s, 7 samples)
Execl Throughput                               3613.2 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        544868.4 KBps  (30.2 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          142354.0 KBps  (30.2 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1789046.2 KBps  (30.2 s, 2 samples)
Pipe Throughput                              799976.2 lps   (10.1 s, 7 samples)
Pipe-based Context Switching                 118596.6 lps   (10.1 s, 7 samples)
Process Creation                               8878.8 lps   (30.2 s, 2 samples)
Shell Scripts (1 concurrent)                   6676.2 lpm   (60.4 s, 2 samples)
Shell Scripts (8 concurrent)                    880.7 lpm   (60.5 s, 2 samples)
System Call Overhead                         577577.6 lps   (10.1 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   22313053.1   1912.0
Double-Precision Whetstone                       55.0       3909.2    710.8
Execl Throughput                                 43.0       3613.2    840.3
File Copy 1024 bufsize 2000 maxblocks          3960.0     544868.4   1375.9
File Copy 256 bufsize 500 maxblocks            1655.0     142354.0    860.1
File Copy 4096 bufsize 8000 maxblocks          5800.0    1789046.2   3084.6
Pipe Throughput                               12440.0     799976.2    643.1
Pipe-based Context Switching                   4000.0     118596.6    296.5
Process Creation                                126.0       8878.8    704.7
Shell Scripts (1 concurrent)                     42.4       6676.2   1574.6
Shell Scripts (8 concurrent)                      6.0        880.7   1467.8
System Call Overhead                          15000.0     577577.6    385.1
                                                                   ========
System Benchmarks Index Score                                         943.6

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 2.33947 s, 918 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 2.39378 s, 897 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=30560: Thu Aug 29 13:11:45 2019
  read : io=6130.3MB, bw=104620KB/s, iops=26155, runt= 60002msec
    clat (usec): min=1, max=177662, avg=300.58, stdev=607.73
     lat (usec): min=2, max=177662, avg=301.36, stdev=607.78
    clat percentiles (usec):
     |  1.00th=[  145],  5.00th=[  221], 10.00th=[  235], 20.00th=[  249],
     | 30.00th=[  258], 40.00th=[  270], 50.00th=[  278], 60.00th=[  286],
     | 70.00th=[  302], 80.00th=[  322], 90.00th=[  370], 95.00th=[  430],
     | 99.00th=[  636], 99.50th=[  764], 99.90th=[ 1512], 99.95th=[ 2064],
     | 99.99th=[ 8256]
    bw (KB  /s): min= 8008, max=14376, per=12.53%, avg=13104.15, stdev=833.24
    lat (usec) : 2=0.01%, 4=0.40%, 10=0.01%, 20=0.02%, 50=0.01%
    lat (usec) : 100=0.03%, 250=20.40%, 500=76.42%, 750=2.17%, 1000=0.35%
    lat (msec) : 2=0.15%, 4=0.03%, 10=0.02%, 20=0.01%, 50=0.01%
    lat (msec) : 100=0.01%, 250=0.01%
  cpu          : usr=1.52%, sys=7.10%, ctx=1586565, majf=0, minf=74
  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    : total=r=1569353/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=6130.3MB, aggrb=104620KB/s, minb=104620KB/s, maxb=104620KB/s, mint=60002msec, maxt=60002msec

Disk stats (read/write):
  sda: ios=1561126/29, merge=24/25, ticks=398356/28, in_queue=398220, util=98.04%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=30592: Thu Aug 29 13:13:50 2019
  write: io=9877.7MB, bw=168561KB/s, iops=42140, runt= 60006msec
    clat (usec): min=1, max=910027, avg=161.10, stdev=4433.63
     lat (usec): min=2, max=910027, avg=165.07, stdev=4447.87
    clat percentiles (usec):
     |  1.00th=[    2],  5.00th=[    2], 10.00th=[    2], 20.00th=[    3],
     | 30.00th=[    3], 40.00th=[    3], 50.00th=[    3], 60.00th=[    3],
     | 70.00th=[    3], 80.00th=[    4], 90.00th=[    5], 95.00th=[   10],
     | 99.00th=[  181], 99.50th=[ 5280], 99.90th=[40192], 99.95th=[63232],
     | 99.99th=[201728]
    bw (KB  /s): min=    0, max=51352, per=13.49%, avg=22743.45, stdev=11130.39
    lat (usec) : 2=0.01%, 4=74.06%, 10=20.37%, 20=3.75%, 50=0.63%
    lat (usec) : 100=0.12%, 250=0.15%, 500=0.08%, 750=0.03%, 1000=0.02%
    lat (msec) : 2=0.07%, 4=0.13%, 10=0.32%, 20=0.10%, 50=0.09%
    lat (msec) : 100=0.05%, 250=0.02%, 500=0.01%, 750=0.01%, 1000=0.01%
  cpu          : usr=1.01%, sys=3.19%, ctx=81638, majf=0, minf=92
  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    : total=r=0/w=2528667/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=9877.7MB, aggrb=168560KB/s, minb=168560KB/s, maxb=168560KB/s, mint=60006msec, maxt=60006msec

Disk stats (read/write):
  sda: ios=413/1142901, merge=133/14847, ticks=840/6249928, in_queue=6248644, util=98.55%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=30573: Thu Aug 29 13:12:50 2019
  read : io=5910.2MB, bw=100863KB/s, iops=25215, runt= 60002msec
    clat (usec): min=63, max=1022.7K, avg=312.03, stdev=1080.17
     lat (usec): min=64, max=1022.7K, avg=312.80, stdev=1080.20
    clat percentiles (usec):
     |  1.00th=[  159],  5.00th=[  221], 10.00th=[  233], 20.00th=[  247],
     | 30.00th=[  258], 40.00th=[  266], 50.00th=[  274], 60.00th=[  286],
     | 70.00th=[  298], 80.00th=[  318], 90.00th=[  358], 95.00th=[  422],
     | 99.00th=[  956], 99.50th=[ 1640], 99.90th=[ 3536], 99.95th=[ 5344],
     | 99.99th=[41728]
    bw (KB  /s): min= 2078, max=14672, per=12.54%, avg=12651.91, stdev=2302.49
    lat (usec) : 100=0.05%, 250=22.44%, 500=74.80%, 750=1.44%, 1000=0.32%
    lat (msec) : 2=0.57%, 4=0.30%, 10=0.07%, 20=0.01%, 50=0.01%
    lat (msec) : 100=0.01%, 250=0.01%, 2000=0.01%
  cpu          : usr=1.39%, sys=6.50%, ctx=1525666, majf=0, minf=59
  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    : total=r=1512996/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=5910.2MB, aggrb=100863KB/s, minb=100863KB/s, maxb=100863KB/s, mint=60002msec, maxt=60002msec

Disk stats (read/write):
  sda: ios=1512242/34, merge=122/40, ticks=413692/76, in_queue=413416, util=98.66%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=30612: Thu Aug 29 13:14:51 2019
  write: io=603464KB, bw=10056KB/s, iops=2514, runt= 60009msec
    clat (usec): min=191, max=485456, avg=3176.44, stdev=18415.40
     lat (usec): min=191, max=485457, avg=3177.24, stdev=18415.46
    clat percentiles (usec):
     |  1.00th=[  262],  5.00th=[  270], 10.00th=[  274], 20.00th=[  282],
     | 30.00th=[  286], 40.00th=[  294], 50.00th=[  302], 60.00th=[  314],
     | 70.00th=[  338], 80.00th=[  394], 90.00th=[  580], 95.00th=[ 1144],
     | 99.00th=[114176], 99.50th=[132096], 99.90th=[162816], 99.95th=[183296],
     | 99.99th=[358400]
    bw (KB  /s): min=   13, max= 2368, per=12.56%, avg=1262.90, stdev=437.28
    lat (usec) : 250=0.04%, 500=86.90%, 750=5.54%, 1000=1.86%
    lat (msec) : 2=2.57%, 4=0.42%, 10=0.20%, 20=0.01%, 50=0.01%
    lat (msec) : 100=0.92%, 250=1.51%, 500=0.02%
  cpu          : usr=0.17%, sys=0.86%, ctx=306183, majf=0, minf=73
  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    : total=r=0/w=150866/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=603464KB, aggrb=10056KB/s, minb=10056KB/s, maxb=10056KB/s, mint=60009msec, maxt=60009msec

Disk stats (read/write):
  sda: ios=158/150940, merge=217/1990, ticks=344/53016, in_queue=53312, util=87.90%

Speed​​test

Retrieving speedtest.net configuration...
Testing from Limited Liability Company Relcom Group (194.87.238.213)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Orange Business Services, Russia & CIS (Moscow) [1.61 km]: 11.038 ms
Testing download speed................................................................................
Download: 732.49 Mbit/s
Testing upload speed................................................................................................
Upload: 99.68 Mbit/s

Timeweb の仮想サーバー:

安価な仮想サーバーのテスト

  • 装置はサンクトペテルブルク (Tsvetochnaya の Selectel DC) にあります。
  • EasyOne 料金表 (5 GB SSD / 1x 2.4 GHz CPU / 0.5 GB RAM / 5 Mbit/s で月額 45 ルーブル);
  • アカウントごとにこの料金が適用される仮想サーバー XNUMX 台。
  • ポート 25、389、465、587、3389、53413 はブロックされています。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-8-amd64 -- #1 SMP Debian 4.9.130-2 (2018-10-27)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version 2.4.0 (4800.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   17:52:57 up 8 days, 17:42,  1 user,  load average: 0.68, 0.21, 0.07; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Mon Sep 09 2019 17:52:57 - 18:21:22
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       32758977.7 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     4454.0 MWIPS (10.7 s, 7 samples)
Execl Throughput                               3775.4 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        291468.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           81792.1 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        870796.6 KBps  (30.0 s, 2 samples)
Pipe Throughput                              434946.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  98160.9 lps   (10.0 s, 7 samples)
Process Creation                               9429.6 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   6887.4 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    908.3 lpm   (60.0 s, 2 samples)
System Call Overhead                         333855.5 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   32758977.7   2807.1
Double-Precision Whetstone                       55.0       4454.0    809.8
Execl Throughput                                 43.0       3775.4    878.0
File Copy 1024 bufsize 2000 maxblocks          3960.0     291468.4    736.0
File Copy 256 bufsize 500 maxblocks            1655.0      81792.1    494.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     870796.6   1501.4
Pipe Throughput                               12440.0     434946.1    349.6
Pipe-based Context Switching                   4000.0      98160.9    245.4
Process Creation                                126.0       9429.6    748.4
Shell Scripts (1 concurrent)                     42.4       6887.4   1624.4
Shell Scripts (8 concurrent)                      6.0        908.3   1513.8
System Call Overhead                          15000.0     333855.5    222.6
                                                                   ========
System Benchmarks Index Score                                         762.0

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 7.336 s, 293 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 3.92267 s, 547 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=12420: Mon Sep  9 18:22:40 2019
  read : io=2487.7MB, bw=42452KB/s, iops=10613, runt= 60005msec
    clat (usec): min=3, max=107480, avg=720.93, stdev=2037.66
     lat (usec): min=4, max=107481, avg=726.16, stdev=2054.61
    clat percentiles (usec):
     |  1.00th=[   47],  5.00th=[   73], 10.00th=[   94], 20.00th=[  133],
     | 30.00th=[  181], 40.00th=[  258], 50.00th=[  314], 60.00th=[  410],
     | 70.00th=[  564], 80.00th=[  844], 90.00th=[ 1224], 95.00th=[ 2160],
     | 99.00th=[ 7200], 99.50th=[11584], 99.90th=[30592], 99.95th=[38144],
     | 99.99th=[57600]
    bw (KB  /s): min= 1176, max=14304, per=12.52%, avg=5315.30, stdev=2443.62
    lat (usec) : 4=0.13%, 10=0.75%, 20=0.02%, 50=0.15%, 100=10.51%
    lat (usec) : 250=27.08%, 500=28.01%, 750=10.66%, 1000=6.97%
    lat (msec) : 2=9.93%, 4=3.57%, 10=1.56%, 20=0.41%, 50=0.21%
    lat (msec) : 100=0.02%, 250=0.01%
  cpu          : usr=2.06%, sys=6.20%, ctx=638836, majf=0, minf=83
  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    : total=r=636840/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=2487.7MB, aggrb=42452KB/s, minb=42452KB/s, maxb=42452KB/s, mint=60005msec, maxt=60005msec

Disk stats (read/write):
  vda: ios=631677/461, merge=17/646, ticks=237832/1172, in_queue=238608, util=85.61%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=12460: Mon Sep  9 18:24:45 2019
  write: io=621160KB, bw=10343KB/s, iops=2585, runt= 60054msec
    clat (usec): min=3, max=1288.9K, avg=2410.29, stdev=27207.90
     lat (usec): min=4, max=1288.1K, avg=2415.84, stdev=27232.16
    clat percentiles (usec):
     |  1.00th=[    4],  5.00th=[    4], 10.00th=[    5], 20.00th=[    5],
     | 30.00th=[    5], 40.00th=[    5], 50.00th=[    6], 60.00th=[    6],
     | 70.00th=[    7], 80.00th=[    9], 90.00th=[   11], 95.00th=[   20],
     | 99.00th=[58112], 99.50th=[199680], 99.90th=[436224], 99.95th=[536576],
     | 99.99th=[634880]
    bw (KB  /s): min=    0, max=42357, per=16.20%, avg=1675.86, stdev=3143.18
    lat (usec) : 4=0.09%, 10=84.01%, 20=10.77%, 50=1.25%, 100=0.96%
    lat (usec) : 250=0.62%, 500=0.31%, 750=0.11%, 1000=0.08%
    lat (msec) : 2=0.07%, 4=0.10%, 10=0.16%, 20=0.16%, 50=0.24%
    lat (msec) : 100=0.35%, 250=0.37%, 500=0.27%, 750=0.06%, 1000=0.01%
    lat (msec) : 2000=0.01%
  cpu          : usr=0.29%, sys=0.75%, ctx=44591, majf=0, minf=69
  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    : total=r=0/w=155290/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=621160KB, aggrb=10343KB/s, minb=10343KB/s, maxb=10343KB/s, mint=60054msec, maxt=60054msec

Disk stats (read/write):
  vda: ios=772/52683, merge=0/13854, ticks=12072/2440176, in_queue=2456616, util=97.18%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=12445: Mon Sep  9 18:23:45 2019
  read : io=3069.2MB, bw=52374KB/s, iops=13093, runt= 60004msec
    clat (usec): min=28, max=192693, avg=581.86, stdev=1688.07
     lat (usec): min=29, max=192695, avg=587.38, stdev=1711.01
    clat percentiles (usec):
     |  1.00th=[   47],  5.00th=[   65], 10.00th=[   83], 20.00th=[  119],
     | 30.00th=[  161], 40.00th=[  239], 50.00th=[  282], 60.00th=[  354],
     | 70.00th=[  474], 80.00th=[  692], 90.00th=[ 1048], 95.00th=[ 1640],
     | 99.00th=[ 5024], 99.50th=[ 8256], 99.90th=[24192], 99.95th=[33536],
     | 99.99th=[60160]
    bw (KB  /s): min=  590, max=19960, per=12.48%, avg=6537.92, stdev=2924.59
    lat (usec) : 50=1.71%, 100=12.40%, 250=29.08%, 500=28.51%, 750=10.21%
    lat (usec) : 1000=6.31%
    lat (msec) : 2=7.87%, 4=2.54%, 10=1.00%, 20=0.25%, 50=0.12%
    lat (msec) : 100=0.02%, 250=0.01%
  cpu          : usr=2.43%, sys=5.97%, ctx=797027, majf=0, minf=59
  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    : total=r=785667/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=3069.2MB, aggrb=52374KB/s, minb=52374KB/s, maxb=52374KB/s, mint=60004msec, maxt=60004msec

Disk stats (read/write):
  vda: ios=785616/21, merge=7/19, ticks=240480/72, in_queue=240176, util=85.74%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=12485: Mon Sep  9 18:25:53 2019
  write: io=194840KB, bw=3246.3KB/s, iops=811, runt= 60021msec
    clat (usec): min=331, max=485350, avg=9837.11, stdev=33348.07
     lat (usec): min=332, max=485351, avg=9839.49, stdev=33349.20
    clat percentiles (usec):
     |  1.00th=[  418],  5.00th=[  446], 10.00th=[  470], 20.00th=[  498],
     | 30.00th=[  532], 40.00th=[  572], 50.00th=[  628], 60.00th=[  740],
     | 70.00th=[  964], 80.00th=[ 1624], 90.00th=[ 5216], 95.00th=[94720],
     | 99.00th=[164864], 99.50th=[195584], 99.90th=[280576], 99.95th=[329728],
     | 99.99th=[403456]
    bw (KB  /s): min=   13, max= 1112, per=12.67%, avg=411.37, stdev=183.33
    lat (usec) : 500=20.24%, 750=40.58%, 1000=10.11%
    lat (msec) : 2=11.58%, 4=6.13%, 10=3.13%, 20=0.61%, 50=0.30%
    lat (msec) : 100=2.91%, 250=4.25%, 500=0.15%
  cpu          : usr=0.20%, sys=1.08%, ctx=101847, majf=0, minf=74
  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    : total=r=0/w=48710/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=194840KB, aggrb=3246KB/s, minb=3246KB/s, maxb=3246KB/s, mint=60021msec, maxt=60021msec

Disk stats (read/write):
  vda: ios=587/48721, merge=7/1768, ticks=1004/54804, in_queue=55760, util=90.21%

Speed​​test

Retrieving speedtest.net configuration...
Testing from TimeWeb Ltd. (5.23.49.159)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Megafon (Moscow) [1.61 km]: 13.918 ms
Testing download speed................................................................................
Download: 3.82 Mbit/s
Testing upload speed................................................................................................
Upload: 4.79 Mbit/s

Inoventica Services の仮想サーバー:

安価な仮想サーバーのテスト

  • Luna プロモーション料金 (5 GB HDD / 1x 2 GHz CPU / 0.5 GB RAM / 10 Mbit/s で月額 49 ルーブル);
  • アカウントごとにこの料金が適用される仮想サーバー XNUMX 台。
  • ポート 25 がブロックされています。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-9-amd64 -- #1 SMP Debian 4.9.168-1+deb9u5 (2019-08-11)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version 2.5+ (3990.4 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   18:50:57 up 11 min,  1 user,  load average: 0.76, 0.28, 0.10; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Thu Aug 29 2019 18:50:57 - 19:18:55
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       18807123.2 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2427.1 MWIPS (9.7 s, 7 samples)
Execl Throughput                               1745.2 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        131818.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           35944.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        424330.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              215841.6 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  44367.6 lps   (10.0 s, 7 samples)
Process Creation                               3968.7 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   3353.6 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    464.0 lpm   (60.1 s, 2 samples)
System Call Overhead                         169640.3 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   18807123.2   1611.6
Double-Precision Whetstone                       55.0       2427.1    441.3
Execl Throughput                                 43.0       1745.2    405.9
File Copy 1024 bufsize 2000 maxblocks          3960.0     131818.8    332.9
File Copy 256 bufsize 500 maxblocks            1655.0      35944.4    217.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     424330.5    731.6
Pipe Throughput                               12440.0     215841.6    173.5
Pipe-based Context Switching                   4000.0      44367.6    110.9
Process Creation                                126.0       3968.7    315.0
Shell Scripts (1 concurrent)                     42.4       3353.6    791.0
Shell Scripts (8 concurrent)                      6.0        464.0    773.3
System Call Overhead                          15000.0     169640.3    113.1
                                                                   ========
System Benchmarks Index Score                                         369.1

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 13.8382 s, 155 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 11.9849 s, 179 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=6399: Thu Aug 29 19:20:36 2019
  read : io=3303.6MB, bw=56376KB/s, iops=14093, runt= 60004msec
    clat (usec): min=5, max=809158, avg=548.42, stdev=1410.87
     lat (usec): min=7, max=809160, avg=551.41, stdev=1412.08
    clat percentiles (usec):
     |  1.00th=[  129],  5.00th=[  233], 10.00th=[  286], 20.00th=[  358],
     | 30.00th=[  414], 40.00th=[  462], 50.00th=[  502], 60.00th=[  548],
     | 70.00th=[  588], 80.00th=[  652], 90.00th=[  764], 95.00th=[  916],
     | 99.00th=[ 1864], 99.50th=[ 2352], 99.90th=[ 4576], 99.95th=[ 6368],
     | 99.99th=[19584]
    bw (KB  /s): min= 2033, max= 9448, per=12.54%, avg=7070.36, stdev=809.28
    lat (usec) : 10=0.54%, 20=0.15%, 50=0.05%, 100=0.03%, 250=5.61%
    lat (usec) : 500=42.50%, 750=40.45%, 1000=6.91%
    lat (msec) : 2=2.92%, 4=0.71%, 10=0.11%, 20=0.02%, 50=0.01%
    lat (msec) : 100=0.01%, 250=0.01%, 750=0.01%, 1000=0.01%
  cpu          : usr=2.97%, sys=8.85%, ctx=850919, majf=0, minf=68
  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    : total=r=845695/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=3303.6MB, aggrb=56375KB/s, minb=56375KB/s, maxb=56375KB/s, mint=60004msec, maxt=60004msec

Disk stats (read/write):
  vda: ios=839215/10, merge=0/3, ticks=110672/32, in_queue=110024, util=94.78%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=6441: Thu Aug 29 19:22:47 2019
  write: io=7166.5MB, bw=122273KB/s, iops=30568, runt= 60017msec
    clat (usec): min=6, max=179576, avg=171.53, stdev=1724.15
     lat (usec): min=7, max=179594, avg=187.29, stdev=1836.72
    clat percentiles (usec):
     |  1.00th=[    7],  5.00th=[    7], 10.00th=[    8], 20.00th=[    8],
     | 30.00th=[    8], 40.00th=[    9], 50.00th=[   10], 60.00th=[   12],
     | 70.00th=[   15], 80.00th=[   17], 90.00th=[   23], 95.00th=[   29],
     | 99.00th=[ 4256], 99.50th=[12224], 99.90th=[24960], 99.95th=[29568],
     | 99.99th=[41216]
    bw (KB  /s): min=    0, max=22003, per=12.65%, avg=15468.37, stdev=3032.64
    lat (usec) : 10=44.94%, 20=41.47%, 50=11.14%, 100=0.37%, 250=0.36%
    lat (usec) : 500=0.13%, 750=0.08%, 1000=0.07%
    lat (msec) : 2=0.18%, 4=0.20%, 10=0.42%, 20=0.43%, 50=0.21%
    lat (msec) : 100=0.01%, 250=0.01%
  cpu          : usr=3.67%, sys=6.27%, ctx=84956, majf=0, minf=67
  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    : total=r=0/w=1834611/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=7166.5MB, aggrb=122272KB/s, minb=122272KB/s, maxb=122272KB/s, mint=60017msec, maxt=60017msec

Disk stats (read/write):
  vda: ios=259/1088561, merge=0/13972, ticks=640/2457092, in_queue=2457680, util=45.40%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=6420: Thu Aug 29 19:21:46 2019
  read : io=3922.6MB, bw=66941KB/s, iops=16735, runt= 60004msec
    clat (usec): min=56, max=56057, avg=456.83, stdev=313.95
     lat (usec): min=59, max=56061, avg=459.82, stdev=315.28
    clat percentiles (usec):
     |  1.00th=[  139],  5.00th=[  197], 10.00th=[  237], 20.00th=[  294],
     | 30.00th=[  338], 40.00th=[  382], 50.00th=[  426], 60.00th=[  470],
     | 70.00th=[  502], 80.00th=[  548], 90.00th=[  644], 95.00th=[  780],
     | 99.00th=[ 1480], 99.50th=[ 2096], 99.90th=[ 3792], 99.95th=[ 4768],
     | 99.99th=[ 7904]
    bw (KB  /s): min= 6408, max=10008, per=12.50%, avg=8369.21, stdev=613.88
    lat (usec) : 100=0.08%, 250=12.02%, 500=57.76%, 750=24.39%, 1000=3.39%
    lat (msec) : 2=1.80%, 4=0.47%, 10=0.08%, 20=0.01%, 50=0.01%
    lat (msec) : 100=0.01%
  cpu          : usr=3.75%, sys=8.16%, ctx=1012596, majf=0, minf=74
  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    : total=r=1004176/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=3922.6MB, aggrb=66940KB/s, minb=66940KB/s, maxb=66940KB/s, mint=60004msec, maxt=60004msec

Disk stats (read/write):
  vda: ios=1004393/8, merge=0/4, ticks=122236/0, in_queue=121560, util=97.14%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=6458: Thu Aug 29 19:23:48 2019
  write: io=832520KB, bw=13873KB/s, iops=3468, runt= 60008msec
    clat (usec): min=85, max=58011, avg=2287.70, stdev=7287.28
     lat (usec): min=87, max=58014, avg=2290.95, stdev=7287.68
    clat percentiles (usec):
     |  1.00th=[  116],  5.00th=[  135], 10.00th=[  145], 20.00th=[  161],
     | 30.00th=[  179], 40.00th=[  197], 50.00th=[  219], 60.00th=[  245],
     | 70.00th=[  282], 80.00th=[  354], 90.00th=[  772], 95.00th=[25216],
     | 99.00th=[33024], 99.50th=[35584], 99.90th=[41216], 99.95th=[44288],
     | 99.99th=[54016]
    bw (KB  /s): min= 1040, max= 2465, per=12.51%, avg=1734.85, stdev=241.51
    lat (usec) : 100=0.06%, 250=61.34%, 500=25.50%, 750=2.98%, 1000=1.04%
    lat (msec) : 2=1.25%, 4=0.45%, 10=0.06%, 20=0.24%, 50=7.08%
    lat (msec) : 100=0.02%
  cpu          : usr=0.85%, sys=3.65%, ctx=432909, majf=0, minf=74
  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    : total=r=0/w=208130/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=832520KB, aggrb=13873KB/s, minb=13873KB/s, maxb=13873KB/s, mint=60008msec, maxt=60008msec

Disk stats (read/write):
  vda: ios=163/208125, merge=0/2312, ticks=104/42948, in_queue=42804, util=69.09%

Speed​​test

Retrieving speedtest.net configuration...
Testing from Unknown (45.143.136.58)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Telstra (Perth) [1840.76 km]: 532.274 ms
Testing download speed................................................................................
Download: 14.41 Mbit/s
Testing upload speed................................................................................................
Upload: 4.44 Mbit/s

Hosting-Russia の仮想サーバー:

安価な仮想サーバーのテスト

  • 社会料金 (10 GB SSD / 1x 3.3 GHz CPU / 0.5 GB RAM / 100 Mbit/s、月額 50 ルーブル);
  • アカウントごとにこの料金の仮想サーバー XNUMX 台。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-9-amd64 -- #1 SMP Debian 4.9.168-1+deb9u5 (2019-08-11)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version 1.5.3 (6600.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   18:48:30 up 8 min,  1 user,  load average: 0.59, 0.20, 0.07; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Thu Aug 29 2019 18:48:30 - 19:16:42
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       32415885.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     4506.4 MWIPS (9.8 s, 7 samples)
Execl Throughput                               2793.0 lps   (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        187339.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           44305.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        520367.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              283190.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  68876.3 lps   (10.0 s, 7 samples)
Process Creation                               6402.6 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   5276.1 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    652.5 lpm   (60.1 s, 2 samples)
System Call Overhead                         225827.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   32415885.5   2777.7
Double-Precision Whetstone                       55.0       4506.4    819.3
Execl Throughput                                 43.0       2793.0    649.5
File Copy 1024 bufsize 2000 maxblocks          3960.0     187339.0    473.1
File Copy 256 bufsize 500 maxblocks            1655.0      44305.5    267.7
File Copy 4096 bufsize 8000 maxblocks          5800.0     520367.5    897.2
Pipe Throughput                               12440.0     283190.8    227.6
Pipe-based Context Switching                   4000.0      68876.3    172.2
Process Creation                                126.0       6402.6    508.1
Shell Scripts (1 concurrent)                     42.4       5276.1   1244.4
Shell Scripts (8 concurrent)                      6.0        652.5   1087.6
System Call Overhead                          15000.0     225827.6    150.6
                                                                   ========
System Benchmarks Index Score                                         544.7

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 8.07309 s, 266 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 9.44878 s, 227 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=838: Thu Aug 29 19:18:10 2019
  read : io=4574.7MB, bw=78068KB/s, iops=19517, runt= 60004msec
    clat (usec): min=4, max=48426, avg=396.43, stdev=500.67
     lat (usec): min=5, max=48428, avg=398.64, stdev=502.34
    clat percentiles (usec):
     |  1.00th=[   30],  5.00th=[  145], 10.00th=[  181], 20.00th=[  223],
     | 30.00th=[  255], 40.00th=[  278], 50.00th=[  302], 60.00th=[  330],
     | 70.00th=[  362], 80.00th=[  426], 90.00th=[  620], 95.00th=[  932],
     | 99.00th=[ 2128], 99.50th=[ 3024], 99.90th=[ 6112], 99.95th=[ 8256],
     | 99.99th=[15424]
    bw (KB  /s): min= 4335, max=14936, per=12.51%, avg=9763.49, stdev=1977.64
    lat (usec) : 10=0.34%, 20=0.51%, 50=0.22%, 100=0.64%, 250=26.50%
    lat (usec) : 500=57.31%, 750=7.19%, 1000=2.86%
    lat (msec) : 2=3.28%, 4=0.86%, 10=0.25%, 20=0.03%, 50=0.01%
  cpu          : usr=2.97%, sys=7.76%, ctx=1166714, majf=0, minf=66
  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    : total=r=1171105/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=4574.7MB, aggrb=78068KB/s, minb=78068KB/s, maxb=78068KB/s, mint=60004msec, maxt=60004msec

Disk stats (read/write):
  vda: ios=1153842/15, merge=0/7, ticks=183000/0, in_queue=182384, util=93.31%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=874: Thu Aug 29 19:20:21 2019
  write: io=10509MB, bw=178979KB/s, iops=44744, runt= 60123msec
    clat (usec): min=4, max=472089, avg=123.05, stdev=2340.78
     lat (usec): min=5, max=472091, avg=133.56, stdev=2395.07
    clat percentiles (usec):
     |  1.00th=[    5],  5.00th=[    5], 10.00th=[    5], 20.00th=[    6],
     | 30.00th=[    6], 40.00th=[    6], 50.00th=[    8], 60.00th=[    9],
     | 70.00th=[    9], 80.00th=[   10], 90.00th=[   14], 95.00th=[   18],
     | 99.00th=[  884], 99.50th=[ 8032], 99.90th=[20352], 99.95th=[25728],
     | 99.99th=[48384]
    bw (KB  /s): min=    0, max=32906, per=12.74%, avg=22795.95, stdev=6415.87
    lat (usec) : 10=72.98%, 20=23.63%, 50=1.87%, 100=0.22%, 250=0.14%
    lat (usec) : 500=0.08%, 750=0.06%, 1000=0.03%
    lat (msec) : 2=0.08%, 4=0.14%, 10=0.35%, 20=0.29%, 50=0.11%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%
  cpu          : usr=3.98%, sys=5.95%, ctx=84003, majf=0, minf=76
  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    : total=r=0/w=2690190/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=10509MB, aggrb=178979KB/s, minb=178979KB/s, maxb=178979KB/s, mint=60123msec, maxt=60123msec

Disk stats (read/write):
  vda: ios=190/1476042, merge=0/13939, ticks=476/4394496, in_queue=4455180, util=69.07%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=857: Thu Aug 29 19:19:21 2019
  read : io=5403.6MB, bw=92212KB/s, iops=23053, runt= 60005msec
    clat (usec): min=39, max=50176, avg=333.70, stdev=389.36
     lat (usec): min=41, max=50178, avg=335.87, stdev=392.12
    clat percentiles (usec):
     |  1.00th=[   99],  5.00th=[  141], 10.00th=[  169], 20.00th=[  205],
     | 30.00th=[  235], 40.00th=[  251], 50.00th=[  274], 60.00th=[  286],
     | 70.00th=[  310], 80.00th=[  350], 90.00th=[  482], 95.00th=[  724],
     | 99.00th=[ 1592], 99.50th=[ 2192], 99.90th=[ 5088], 99.95th=[ 6560],
     | 99.99th=[11840]
    bw (KB  /s): min= 4031, max=16680, per=12.49%, avg=11517.12, stdev=2314.89
    lat (usec) : 50=0.01%, 100=1.01%, 250=38.14%, 500=51.46%, 750=4.69%
    lat (usec) : 1000=2.01%
    lat (msec) : 2=2.08%, 4=0.46%, 10=0.14%, 20=0.01%, 50=0.01%
    lat (msec) : 100=0.01%
  cpu          : usr=3.41%, sys=7.62%, ctx=1389080, majf=0, minf=74
  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    : total=r=1383298/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=5403.6MB, aggrb=92212KB/s, minb=92212KB/s, maxb=92212KB/s, mint=60005msec, maxt=60005msec

Disk stats (read/write):
  vda: ios=1381796/8, merge=0/4, ticks=184404/0, in_queue=183664, util=95.35%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=891: Thu Aug 29 19:21:22 2019
  write: io=1274.2MB, bw=21758KB/s, iops=5439, runt= 60006msec
    clat (usec): min=53, max=313006, avg=1458.70, stdev=7507.36
     lat (usec): min=54, max=313008, avg=1460.81, stdev=7507.98
    clat percentiles (usec):
     |  1.00th=[   67],  5.00th=[   76], 10.00th=[   81], 20.00th=[   87],
     | 30.00th=[   94], 40.00th=[  103], 50.00th=[  114], 60.00th=[  129],
     | 70.00th=[  149], 80.00th=[  183], 90.00th=[  290], 95.00th=[ 1128],
     | 99.00th=[32384], 99.50th=[37120], 99.90th=[49408], 99.95th=[59136],
     | 99.99th=[264192]
    bw (KB  /s): min=  686, max= 5032, per=12.62%, avg=2744.87, stdev=691.50
    lat (usec) : 100=36.39%, 250=51.54%, 500=5.55%, 750=0.89%, 1000=0.46%
    lat (msec) : 2=0.59%, 4=0.24%, 10=0.12%, 20=0.10%, 50=4.02%
    lat (msec) : 100=0.06%, 250=0.01%, 500=0.03%
  cpu          : usr=0.84%, sys=3.25%, ctx=663224, majf=0, minf=83
  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    : total=r=0/w=326399/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=1274.2MB, aggrb=21757KB/s, minb=21757KB/s, maxb=21757KB/s, mint=60006msec, maxt=60006msec

Disk stats (read/write):
  vda: ios=137/326331, merge=0/2139, ticks=132/47148, in_queue=47096, util=73.67%

Speed​​test

Retrieving speedtest.net configuration...
Testing from Argotel ANTIDDOS-HOSTING (95.182.120.105)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Orange Business Services, Russia & CIS (Moscow) [0.12 km]: 5.474 ms
Testing download speed................................................................................
Download: 104.50 Mbit/s
Testing upload speed................................................................................................
Upload: 118.08 Mbit/s

FirstByte の仮想サーバー:

安価な仮想サーバーのテスト

  • 料金表 MSK-KVM-SSD-START (7 GB SSD / 1x 2.9 GHz CPU / 0.5 GB RAM / 100 Mbit/s、7 TB トラフィック、月額 55 ルーブル)。
  • アカウントごとにこの料金で XNUMX つの仮想サーバーが適用されます。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-9-amd64 -- #1 SMP Debian 4.9.168-1+deb9u5 (2019-08-11)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version 2.5+ (5800.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   18:51:35 up  3:21,  1 user,  load average: 0.59, 0.17, 0.06; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Fri Aug 30 2019 18:51:35 - 19:19:41
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       35010835.0 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     4233.8 MWIPS (9.9 s, 7 samples)
Execl Throughput                               3115.3 lps   (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        230064.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           61011.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        749773.8 KBps  (30.0 s, 2 samples)
Pipe Throughput                              315333.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  69505.3 lps   (10.0 s, 7 samples)
Process Creation                               7651.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   6090.7 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    812.0 lpm   (60.0 s, 2 samples)
System Call Overhead                         275532.7 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   35010835.0   3000.1
Double-Precision Whetstone                       55.0       4233.8    769.8
Execl Throughput                                 43.0       3115.3    724.5
File Copy 1024 bufsize 2000 maxblocks          3960.0     230064.0    581.0
File Copy 256 bufsize 500 maxblocks            1655.0      61011.7    368.7
File Copy 4096 bufsize 8000 maxblocks          5800.0     749773.8   1292.7
Pipe Throughput                               12440.0     315333.8    253.5
Pipe-based Context Switching                   4000.0      69505.3    173.8
Process Creation                                126.0       7651.1    607.2
Shell Scripts (1 concurrent)                     42.4       6090.7   1436.5
Shell Scripts (8 concurrent)                      6.0        812.0   1353.4
System Call Overhead                          15000.0     275532.7    183.7
                                                                   ========
System Benchmarks Index Score                                         636.5

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 7.39888 s, 290 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 7.95401 s, 270 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=7673: Fri Aug 30 19:21:06 2019
  read : io=5185.1MB, bw=88488KB/s, iops=22121, runt= 60003msec
    clat (usec): min=3, max=177882, avg=349.83, stdev=464.85
     lat (usec): min=5, max=177883, avg=351.69, stdev=465.74
    clat percentiles (usec):
     |  1.00th=[   32],  5.00th=[  183], 10.00th=[  215], 20.00th=[  249],
     | 30.00th=[  274], 40.00th=[  294], 50.00th=[  314], 60.00th=[  342],
     | 70.00th=[  374], 80.00th=[  418], 90.00th=[  490], 95.00th=[  564],
     | 99.00th=[  828], 99.50th=[ 1416], 99.90th=[ 3024], 99.95th=[ 5280],
     | 99.99th=[12096]
    bw (KB  /s): min= 6592, max=14160, per=12.50%, avg=11059.80, stdev=1182.28
    lat (usec) : 4=0.01%, 10=0.18%, 20=0.61%, 50=0.25%, 100=0.16%
    lat (usec) : 250=19.30%, 500=70.27%, 750=7.90%, 1000=0.62%
    lat (msec) : 2=0.49%, 4=0.15%, 10=0.05%, 20=0.01%, 50=0.01%
    lat (msec) : 100=0.01%, 250=0.01%
  cpu          : usr=2.69%, sys=9.33%, ctx=1328724, majf=0, minf=75
  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    : total=r=1327383/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=5185.1MB, aggrb=88487KB/s, minb=88487KB/s, maxb=88487KB/s, mint=60003msec, maxt=60003msec

Disk stats (read/write):
  vda: ios=1309181/16, merge=0/10, ticks=79784/0, in_queue=79032, util=92.41%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=7711: Fri Aug 30 19:23:16 2019
  write: io=11638MB, bw=198581KB/s, iops=49645, runt= 60012msec
    clat (usec): min=4, max=60776, avg=105.51, stdev=1311.60
     lat (usec): min=5, max=60778, avg=115.51, stdev=1405.26
    clat percentiles (usec):
     |  1.00th=[    5],  5.00th=[    5], 10.00th=[    5], 20.00th=[    6],
     | 30.00th=[    6], 40.00th=[    6], 50.00th=[    7], 60.00th=[    7],
     | 70.00th=[   10], 80.00th=[   12], 90.00th=[   13], 95.00th=[   16],
     | 99.00th=[  636], 99.50th=[ 7200], 99.90th=[21888], 99.95th=[28032],
     | 99.99th=[39168]
    bw (KB  /s): min=    1, max=32893, per=12.62%, avg=25065.22, stdev=3735.85
    lat (usec) : 10=69.83%, 20=27.13%, 50=1.60%, 100=0.20%, 250=0.13%
    lat (usec) : 500=0.07%, 750=0.11%, 1000=0.06%
    lat (msec) : 2=0.11%, 4=0.12%, 10=0.26%, 20=0.26%, 50=0.13%
    lat (msec) : 100=0.01%
  cpu          : usr=3.80%, sys=6.44%, ctx=103619, majf=0, minf=75
  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    : total=r=0/w=2979313/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=11638MB, aggrb=198581KB/s, minb=198581KB/s, maxb=198581KB/s, mint=60012msec, maxt=60012msec

Disk stats (read/write):
  vda: ios=788/1799427, merge=0/13447, ticks=148/1828308, in_queue=1828228, util=40.15%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=7696: Fri Aug 30 19:22:15 2019
  read : io=6190.2MB, bw=105641KB/s, iops=26410, runt= 60003msec
    clat (usec): min=40, max=54082, avg=289.23, stdev=187.94
     lat (usec): min=41, max=54084, avg=291.23, stdev=188.73
    clat percentiles (usec):
     |  1.00th=[  114],  5.00th=[  163], 10.00th=[  189], 20.00th=[  219],
     | 30.00th=[  233], 40.00th=[  249], 50.00th=[  266], 60.00th=[  282],
     | 70.00th=[  314], 80.00th=[  354], 90.00th=[  406], 95.00th=[  458],
     | 99.00th=[  628], 99.50th=[  756], 99.90th=[ 1496], 99.95th=[ 2384],
     | 99.99th=[ 6688]
    bw (KB  /s): min=10160, max=15992, per=12.51%, avg=13220.39, stdev=1164.37
    lat (usec) : 50=0.01%, 100=0.48%, 250=39.69%, 500=56.83%, 750=2.47%
    lat (usec) : 1000=0.28%
    lat (msec) : 2=0.17%, 4=0.04%, 10=0.02%, 20=0.01%, 50=0.01%
    lat (msec) : 100=0.01%
  cpu          : usr=3.74%, sys=8.49%, ctx=1597314, majf=0, minf=74
  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    : total=r=1584690/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=6190.2MB, aggrb=105640KB/s, minb=105640KB/s, maxb=105640KB/s, mint=60003msec, maxt=60003msec

Disk stats (read/write):
  vda: ios=1582870/8, merge=0/6, ticks=92084/0, in_queue=91392, util=97.29%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=7730: Fri Aug 30 19:24:16 2019
  write: io=2561.9MB, bw=43718KB/s, iops=10929, runt= 60006msec
    clat (usec): min=41, max=73708, avg=722.90, stdev=3654.74
     lat (usec): min=42, max=73710, avg=724.50, stdev=3654.92
    clat percentiles (usec):
     |  1.00th=[   52],  5.00th=[   54], 10.00th=[   56], 20.00th=[   60],
     | 30.00th=[   65], 40.00th=[   70], 50.00th=[   75], 60.00th=[   79],
     | 70.00th=[   84], 80.00th=[   92], 90.00th=[  112], 95.00th=[  175],
     | 99.00th=[21632], 99.50th=[22656], 99.90th=[27008], 99.95th=[33024],
     | 99.99th=[46848]
    bw (KB  /s): min= 3652, max= 7496, per=12.50%, avg=5465.98, stdev=541.88
    lat (usec) : 50=0.13%, 100=84.74%, 250=11.09%, 500=0.68%, 750=0.16%
    lat (usec) : 1000=0.05%
    lat (msec) : 2=0.04%, 4=0.01%, 10=0.01%, 20=1.17%, 50=1.90%
    lat (msec) : 100=0.01%
  cpu          : usr=1.20%, sys=4.47%, ctx=1334127, majf=0, minf=66
  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    : total=r=0/w=655834/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=2561.9MB, aggrb=43717KB/s, minb=43717KB/s, maxb=43717KB/s, mint=60006msec, maxt=60006msec

Disk stats (read/write):
  vda: ios=784/655134, merge=0/1513, ticks=148/41460, in_queue=41388, util=64.67%

Speed​​test

Retrieving speedtest.net configuration...
Testing from NTX Technologies s.r.o. (147.78.65.137)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Megafon (Moscow) [1.61 km]: 23.363 ms
Testing download speed................................................................................
Download: 87.97 Mbit/s
Testing upload speed................................................................................................
Upload: 103.26 Mbit/s

VDSina からの仮想サーバー:

安価な仮想サーバーのテスト

  • 機器はロシア (モスクワ、Aviamotornaya の DataPro DC) とオランダ (メッペル、Servius DC) にあります。
  • 2 日あたり 5 ルーブルの名前のない料金 (1 GB SSD / 2.1x 0.5 GHz CPU / 100 GB RAM / 1 Mbit/s、59 TB トラフィックで月額 XNUMX ルーブル)。
  • アカウントごとにこの料金の仮想サーバー XNUMX 台。

テスト:

Unixベンチ (RU)

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-6-amd64 -- #1 SMP Debian 4.9.88-1+deb9u1 (2018-05-07)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Common KVM processor (4394.9 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   13:05:33 up 205 days,  2:51,  1 user,  load average: 0.60, 0.17, 0.06; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Fri Aug 09 2019 13:05:33 - 13:33:38
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       22416507.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3992.1 MWIPS (9.6 s, 7 samples)
Execl Throughput                               2605.9 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        219066.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           59081.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        697518.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                              310130.4 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  72366.5 lps   (10.0 s, 7 samples)
Process Creation                               6019.5 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4518.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    577.9 lpm   (60.1 s, 2 samples)
System Call Overhead                         259893.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   22416507.5   1920.9
Double-Precision Whetstone                       55.0       3992.1    725.8
Execl Throughput                                 43.0       2605.9    606.0
File Copy 1024 bufsize 2000 maxblocks          3960.0     219066.0    553.2
File Copy 256 bufsize 500 maxblocks            1655.0      59081.7    357.0
File Copy 4096 bufsize 8000 maxblocks          5800.0     697518.9   1202.6
Pipe Throughput                               12440.0     310130.4    249.3
Pipe-based Context Switching                   4000.0      72366.5    180.9
Process Creation                                126.0       6019.5    477.7
Shell Scripts (1 concurrent)                     42.4       4518.9   1065.8
Shell Scripts (8 concurrent)                      6.0        577.9    963.2
System Call Overhead                          15000.0     259893.1    173.3
                                                                   ========
System Benchmarks Index Score                                         550.2

Unixベンチ (NL)

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-6-amd64 -- #1 SMP Debian 4.9.88-1+deb9u1 (2018-05-07)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Common KVM processor (4200.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   22:06:14 up 227 days, 11:55,  1 user,  load average: 0.69, 0.21, 0.08; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Sat Aug 31 2019 22:06:14 - 22:34:27
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       22795602.9 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3167.3 MWIPS (9.9 s, 7 samples)
Execl Throughput                               2537.0 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        206972.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           52857.1 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        658303.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              292297.5 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  68596.0 lps   (10.0 s, 7 samples)
Process Creation                               6200.3 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4572.0 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    571.9 lpm   (60.1 s, 2 samples)
System Call Overhead                         240979.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   22795602.9   1953.4
Double-Precision Whetstone                       55.0       3167.3    575.9
Execl Throughput                                 43.0       2537.0    590.0
File Copy 1024 bufsize 2000 maxblocks          3960.0     206972.0    522.7
File Copy 256 bufsize 500 maxblocks            1655.0      52857.1    319.4
File Copy 4096 bufsize 8000 maxblocks          5800.0     658303.5   1135.0
Pipe Throughput                               12440.0     292297.5    235.0
Pipe-based Context Switching                   4000.0      68596.0    171.5
Process Creation                                126.0       6200.3    492.1
Shell Scripts (1 concurrent)                     42.4       4572.0   1078.3
Shell Scripts (8 concurrent)                      6.0        571.9    953.1
System Call Overhead                          15000.0     240979.1    160.7
                                                                   ========
System Benchmarks Index Score                                         522.3

dd (RU)1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 4.02597 s, 533 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 4.766 s, 451 MB/s

dd (オランダ語)1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 3.70351 s, 580 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 4.03231 s, 533 MB/s

フィオ (RU)読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=14888: Fri Aug  9 13:34:53 2019
  read : io=4741.6MB, bw=80914KB/s, iops=20228, runt= 60006msec
    clat (usec): min=4, max=77875, avg=383.55, stdev=631.08
     lat (usec): min=5, max=77877, avg=385.67, stdev=632.08
    clat percentiles (usec):
     |  1.00th=[   25],  5.00th=[  175], 10.00th=[  207], 20.00th=[  227],
     | 30.00th=[  249], 40.00th=[  270], 50.00th=[  294], 60.00th=[  318],
     | 70.00th=[  354], 80.00th=[  414], 90.00th=[  548], 95.00th=[  764],
     | 99.00th=[ 1848], 99.50th=[ 2992], 99.90th=[ 8256], 99.95th=[12096],
     | 99.99th=[23168]
    bw (KB  /s): min= 6360, max=13704, per=12.53%, avg=10139.33, stdev=1257.00
    lat (usec) : 10=0.71%, 20=0.24%, 50=0.09%, 100=0.80%, 250=28.59%
    lat (usec) : 500=57.08%, 750=7.28%, 1000=2.17%
    lat (msec) : 2=2.16%, 4=0.53%, 10=0.27%, 20=0.06%, 50=0.01%
    lat (msec) : 100=0.01%
  cpu          : usr=2.34%, sys=7.79%, ctx=1212630, majf=0, minf=74
  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    : total=r=1213831/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=4741.6MB, aggrb=80913KB/s, minb=80913KB/s, maxb=80913KB/s, mint=60006msec, maxt=60006msec

Disk stats (read/write):
  vda: ios=1198794/8, merge=0/4, ticks=294008/0, in_queue=293620, util=97.13%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=14918: Fri Aug  9 13:36:59 2019
  write: io=11840MB, bw=202049KB/s, iops=50512, runt= 60004msec
    clat (usec): min=3, max=148780, avg=101.62, stdev=1269.68
     lat (usec): min=5, max=148782, avg=111.57, stdev=1359.30
    clat percentiles (usec):
     |  1.00th=[    4],  5.00th=[    5], 10.00th=[    5], 20.00th=[    5],
     | 30.00th=[    6], 40.00th=[    6], 50.00th=[    7], 60.00th=[    7],
     | 70.00th=[    8], 80.00th=[    9], 90.00th=[   12], 95.00th=[   15],
     | 99.00th=[  498], 99.50th=[ 7264], 99.90th=[20096], 99.95th=[25472],
     | 99.99th=[36096]
    bw (KB  /s): min=    0, max=34640, per=12.79%, avg=25851.29, stdev=4081.36
    lat (usec) : 4=0.01%, 10=83.25%, 20=14.37%, 50=0.97%, 100=0.19%
    lat (usec) : 250=0.14%, 500=0.08%, 750=0.08%, 1000=0.05%
    lat (msec) : 2=0.09%, 4=0.12%, 10=0.30%, 20=0.26%, 50=0.11%
    lat (msec) : 100=0.01%, 250=0.01%
  cpu          : usr=3.44%, sys=6.63%, ctx=111831, majf=0, minf=75
  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    : total=r=0/w=3030936/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=11840MB, aggrb=202048KB/s, minb=202048KB/s, maxb=202048KB/s, mint=60004msec, maxt=60004msec

Disk stats (read/write):
  vda: ios=142/1859138, merge=0/13538, ticks=132/2900344, in_queue=2900056, util=50.68%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=14905: Fri Aug  9 13:35:58 2019
  read : io=4862.9MB, bw=82985KB/s, iops=20746, runt= 60005msec
    clat (usec): min=55, max=41916, avg=373.89, stdev=646.76
     lat (usec): min=56, max=41918, avg=375.90, stdev=647.57
    clat percentiles (usec):
     |  1.00th=[   93],  5.00th=[  155], 10.00th=[  201], 20.00th=[  223],
     | 30.00th=[  241], 40.00th=[  262], 50.00th=[  286], 60.00th=[  310],
     | 70.00th=[  346], 80.00th=[  410], 90.00th=[  540], 95.00th=[  724],
     | 99.00th=[ 1688], 99.50th=[ 3088], 99.90th=[ 9152], 99.95th=[13888],
     | 99.99th=[24192]
    bw (KB  /s): min= 5432, max=14842, per=12.51%, avg=10377.91, stdev=1508.66
    lat (usec) : 100=1.51%, 250=33.24%, 500=53.18%, 750=7.51%, 1000=2.23%
    lat (msec) : 2=1.54%, 4=0.42%, 10=0.29%, 20=0.07%, 50=0.02%
  cpu          : usr=2.47%, sys=7.27%, ctx=1256353, majf=0, minf=74
  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    : total=r=1244880/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=4862.9MB, aggrb=82985KB/s, minb=82985KB/s, maxb=82985KB/s, mint=60005msec, maxt=60005msec

Disk stats (read/write):
  vda: ios=1242472/8, merge=0/4, ticks=323432/0, in_queue=322888, util=98.79%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=14931: Fri Aug  9 13:38:00 2019
  write: io=1278.5MB, bw=21815KB/s, iops=5453, runt= 60008msec
    clat (usec): min=73, max=85953, avg=1453.88, stdev=5847.55
     lat (usec): min=74, max=85955, avg=1455.93, stdev=5847.93
    clat percentiles (usec):
     |  1.00th=[   91],  5.00th=[  100], 10.00th=[  107], 20.00th=[  120],
     | 30.00th=[  131], 40.00th=[  139], 50.00th=[  147], 60.00th=[  157],
     | 70.00th=[  171], 80.00th=[  195], 90.00th=[  255], 95.00th=[ 2608],
     | 99.00th=[29824], 99.50th=[32640], 99.90th=[41728], 99.95th=[46336],
     | 99.99th=[62208]
    bw (KB  /s): min= 1799, max= 4007, per=12.51%, avg=2728.25, stdev=332.39
    lat (usec) : 100=4.58%, 250=85.00%, 500=4.55%, 750=0.45%, 1000=0.14%
    lat (msec) : 2=0.23%, 4=0.12%, 10=0.09%, 20=0.50%, 50=4.32%
    lat (msec) : 100=0.03%
  cpu          : usr=0.82%, sys=3.82%, ctx=670541, majf=0, minf=74
  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    : total=r=0/w=327272/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=1278.5MB, aggrb=21815KB/s, minb=21815KB/s, maxb=21815KB/s, mint=60008msec, maxt=60008msec

Disk stats (read/write):
  vda: ios=139/327181, merge=0/2105, ticks=0/42904, in_queue=42740, util=71.21%

フィオ (オランダ)読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=18176: Sat Aug 31 22:35:40 2019
  read : io=5277.5MB, bw=90061KB/s, iops=22515, runt= 60005msec
    clat (usec): min=4, max=29089, avg=343.49, stdev=324.24
     lat (usec): min=5, max=29090, avg=345.52, stdev=324.73
    clat percentiles (usec):
     |  1.00th=[   98],  5.00th=[  151], 10.00th=[  169], 20.00th=[  193],
     | 30.00th=[  211], 40.00th=[  231], 50.00th=[  258], 60.00th=[  298],
     | 70.00th=[  374], 80.00th=[  462], 90.00th=[  580], 95.00th=[  732],
     | 99.00th=[ 1432], 99.50th=[ 1736], 99.90th=[ 2832], 99.95th=[ 3888],
     | 99.99th=[11968]
    bw (KB  /s): min= 5195, max=14704, per=12.50%, avg=11256.88, stdev=1315.93
    lat (usec) : 10=0.52%, 20=0.33%, 50=0.11%, 100=0.05%, 250=46.72%
    lat (usec) : 500=36.35%, 750=11.29%, 1000=2.42%
    lat (msec) : 2=1.92%, 4=0.23%, 10=0.04%, 20=0.01%, 50=0.01%
  cpu          : usr=2.72%, sys=7.50%, ctx=1347767, majf=0, minf=74
  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    : total=r=1351030/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=5277.5MB, aggrb=90061KB/s, minb=90061KB/s, maxb=90061KB/s, mint=60005msec, maxt=60005msec

Disk stats (read/write):
  vda: ios=1335524/8, merge=0/4, ticks=280996/0, in_queue=280604, util=95.44%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=18210: Sat Aug 31 22:37:45 2019
  write: io=11947MB, bw=203870KB/s, iops=50967, runt= 60009msec
    clat (usec): min=4, max=179317, avg=99.95, stdev=1275.88
     lat (usec): min=5, max=179319, avg=110.44, stdev=1364.10
    clat percentiles (usec):
     |  1.00th=[    5],  5.00th=[    5], 10.00th=[    5], 20.00th=[    6],
     | 30.00th=[    6], 40.00th=[    6], 50.00th=[    7], 60.00th=[    7],
     | 70.00th=[    7], 80.00th=[    8], 90.00th=[    9], 95.00th=[   12],
     | 99.00th=[  235], 99.50th=[ 7456], 99.90th=[20352], 99.95th=[24704],
     | 99.99th=[33024]
    bw (KB  /s): min=    0, max=31360, per=12.62%, avg=25722.15, stdev=3838.84
    lat (usec) : 10=90.21%, 20=7.85%, 50=0.68%, 100=0.15%, 250=0.11%
    lat (usec) : 500=0.05%, 750=0.05%, 1000=0.04%
    lat (msec) : 2=0.08%, 4=0.11%, 10=0.27%, 20=0.27%, 50=0.11%
    lat (msec) : 100=0.01%, 250=0.01%
  cpu          : usr=3.82%, sys=6.31%, ctx=83743, majf=0, minf=75
  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    : total=r=0/w=3058507/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=11947MB, aggrb=203869KB/s, minb=203869KB/s, maxb=203869KB/s, mint=60009msec, maxt=60009msec

Disk stats (read/write):
  vda: ios=105/1816322, merge=0/14012, ticks=12/2854440, in_queue=2854096, util=50.34%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=18189: Sat Aug 31 22:36:45 2019
  read : io=6239.2MB, bw=106475KB/s, iops=26618, runt= 60004msec
    clat (usec): min=60, max=22025, avg=289.20, stdev=182.91
     lat (usec): min=61, max=22026, avg=291.16, stdev=183.33
    clat percentiles (usec):
     |  1.00th=[  119],  5.00th=[  143], 10.00th=[  159], 20.00th=[  179],
     | 30.00th=[  197], 40.00th=[  213], 50.00th=[  233], 60.00th=[  262],
     | 70.00th=[  322], 80.00th=[  398], 90.00th=[  486], 95.00th=[  572],
     | 99.00th=[  804], 99.50th=[  948], 99.90th=[ 1576], 99.95th=[ 2040],
     | 99.99th=[ 4384]
    bw (KB  /s): min=10728, max=15296, per=12.51%, avg=13319.46, stdev=800.29
    lat (usec) : 100=0.13%, 250=56.08%, 500=34.90%, 750=7.52%, 1000=0.97%
    lat (msec) : 2=0.35%, 4=0.04%, 10=0.01%, 20=0.01%, 50=0.01%
  cpu          : usr=2.90%, sys=7.37%, ctx=1602606, majf=0, minf=74
  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    : total=r=1597231/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=6239.2MB, aggrb=106474KB/s, minb=106474KB/s, maxb=106474KB/s, mint=60004msec, maxt=60004msec

Disk stats (read/write):
  vda: ios=1596228/8, merge=0/4, ticks=310328/0, in_queue=309840, util=97.91%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=18223: Sat Aug 31 22:38:46 2019
  write: io=940828KB, bw=15677KB/s, iops=3919, runt= 60013msec
    clat (usec): min=63, max=66982, avg=2025.70, stdev=6831.08
     lat (usec): min=64, max=66985, avg=2028.27, stdev=6831.39
    clat percentiles (usec):
     |  1.00th=[   87],  5.00th=[  101], 10.00th=[  113], 20.00th=[  137],
     | 30.00th=[  157], 40.00th=[  177], 50.00th=[  199], 60.00th=[  227],
     | 70.00th=[  278], 80.00th=[  386], 90.00th=[  564], 95.00th=[24960],
     | 99.00th=[31104], 99.50th=[33024], 99.90th=[37120], 99.95th=[39680],
     | 99.99th=[51456]
    bw (KB  /s): min=  924, max= 2897, per=12.51%, avg=1961.38, stdev=295.08
    lat (usec) : 100=4.43%, 250=61.00%, 500=22.51%, 750=4.27%, 1000=0.74%
    lat (msec) : 2=0.41%, 4=0.09%, 10=0.03%, 20=0.07%, 50=6.44%
    lat (msec) : 100=0.01%
  cpu          : usr=0.80%, sys=3.73%, ctx=480789, majf=0, minf=74
  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    : total=r=0/w=235207/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=940828KB, aggrb=15677KB/s, minb=15677KB/s, maxb=15677KB/s, mint=60013msec, maxt=60013msec

Disk stats (read/write):
  vda: ios=107/235136, merge=0/2166, ticks=132/42060, in_queue=42084, util=69.93%

スピードテスト (RU)

Retrieving speedtest.net configuration...
Testing from Hosting technology LTD (185.231.154.182)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Beeline (Moscow) [1.61 km]: 10.909 ms
Testing download speed................................................................................
Download: 178.50 Mbit/s
Testing upload speed................................................................................................
Upload: 231.94 Mbit/s

スピードテスト (NL)

Retrieving speedtest.net configuration...
Testing from McHost.Ru (185.251.38.60)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by HugeServer Networks, LLC (Meppel) [0.74 km]: 4.525 ms
Testing download speed................................................................................
Download: 429.74 Mbit/s
Testing upload speed................................................................................................
Upload: 776.85 Mbit/s

Zomro の仮想サーバー:

安価な仮想サーバーのテスト

  • 機器はオランダ (Dronten、DC Serverius) にあります。
  • 開始料金 (5 GB SSD / 1x 2.13 GHz CPU / 0.25 GB RAM / 10 Mbit/s、月額 0.99 ドル)。
  • 最低レンタル期間は3ヶ月です。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 2.6.32-042stab128.2 -- #1 SMP Thu Mar 22 10:58:36 MSK 2018
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: Intel(R) Xeon(R) CPU L5630 @ 2.13GHz (4267.0 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   10:05:17 up 393 days, 21:23,  1 user,  load average: 1.02, 0.68, 0.31; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Thu Aug 01 2019 10:05:17 - 10:33:49
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       23112316.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2971.7 MWIPS (9.9 s, 7 samples)
Execl Throughput                               2567.9 lps   (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        316297.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           86249.2 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        871594.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                              583546.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 122504.2 lps   (10.0 s, 7 samples)
Process Creation                               4953.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4432.1 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    611.2 lpm   (60.1 s, 2 samples)
System Call Overhead                         434561.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   23112316.5   1980.5
Double-Precision Whetstone                       55.0       2971.7    540.3
Execl Throughput                                 43.0       2567.9    597.2
File Copy 1024 bufsize 2000 maxblocks          3960.0     316297.0    798.7
File Copy 256 bufsize 500 maxblocks            1655.0      86249.2    521.1
File Copy 4096 bufsize 8000 maxblocks          5800.0     871594.9   1502.7
Pipe Throughput                               12440.0     583546.7    469.1
Pipe-based Context Switching                   4000.0     122504.2    306.3
Process Creation                                126.0       4953.8    393.2
Shell Scripts (1 concurrent)                     42.4       4432.1   1045.3
Shell Scripts (8 concurrent)                      6.0        611.2   1018.7
System Call Overhead                          15000.0     434561.8    289.7
                                                                   ========
System Benchmarks Index Score                                         661.1

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 40.3637 s, 53.2 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 38.1745 s, 56.3 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=11077: Thu Aug  1 11:12:41 2019
  read : io=12056KB, bw=205208B/s, iops=50, runt= 60160msec
    clat (msec): min=1, max=281, avg=159.48, stdev=24.08
     lat (msec): min=1, max=281, avg=159.48, stdev=24.08
    clat percentiles (msec):
     |  1.00th=[   21],  5.00th=[  139], 10.00th=[  141], 20.00th=[  159],
     | 30.00th=[  161], 40.00th=[  161], 50.00th=[  161], 60.00th=[  161],
     | 70.00th=[  161], 80.00th=[  161], 90.00th=[  180], 95.00th=[  182],
     | 99.00th=[  221], 99.50th=[  241], 99.90th=[  262], 99.95th=[  281],
     | 99.99th=[  281]
    lat (msec) : 2=0.03%, 20=0.23%, 50=1.06%, 100=0.53%, 250=97.88%
    lat (msec) : 500=0.27%
  cpu          : usr=0.01%, sys=0.03%, ctx=6085, majf=0, minf=298
  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    : total=r=3014/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=12056KB, aggrb=200KB/s, minb=200KB/s, maxb=200KB/s, mint=60160msec, maxt=60160msec

Disk stats (read/write):
  ploop46225p1: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=11110: Thu Aug  1 11:18:57 2019
  write: io=191912KB, bw=3197.5KB/s, iops=799, runt= 60021msec
    clat (usec): min=3, max=36798K, avg=8462.52, stdev=335988.75
     lat (usec): min=3, max=36798K, avg=8463.51, stdev=335988.78
    clat percentiles (usec):
     |  1.00th=[    5],  5.00th=[    5], 10.00th=[    6], 20.00th=[    7],
     | 30.00th=[    8], 40.00th=[    8], 50.00th=[    8], 60.00th=[    9],
     | 70.00th=[   12], 80.00th=[   18], 90.00th=[ 2160], 95.00th=[ 4192],
     | 99.00th=[18304], 99.50th=[140288], 99.90th=[1449984], 99.95th=[1794048],
     | 99.99th=[16711680]
    lat (usec) : 4=0.01%, 10=60.67%, 20=20.72%, 50=3.64%, 100=0.60%
    lat (usec) : 250=0.19%, 500=0.01%, 750=0.02%, 1000=0.38%
    lat (msec) : 2=3.30%, 4=4.49%, 10=4.58%, 20=0.43%, 50=0.05%
    lat (msec) : 100=0.20%, 250=0.44%, 500=0.01%, 750=0.02%, 1000=0.03%
    lat (msec) : 2000=0.18%, >=2000=0.04%
  cpu          : usr=0.04%, sys=0.19%, ctx=18235, majf=0, minf=286
  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    : total=r=0/w=47978/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=191912KB, aggrb=3197KB/s, minb=3197KB/s, maxb=3197KB/s, mint=60021msec, maxt=60021msec

Disk stats (read/write):
  ploop46225p1: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=11090: Thu Aug  1 11:16:30 2019
  read : io=12056KB, bw=205208B/s, iops=50, runt= 60160msec
    clat (msec): min=1, max=281, avg=159.49, stdev=27.57
     lat (msec): min=1, max=281, avg=159.49, stdev=27.57
    clat percentiles (msec):
     |  1.00th=[   21],  5.00th=[  121], 10.00th=[  141], 20.00th=[  159],
     | 30.00th=[  161], 40.00th=[  161], 50.00th=[  161], 60.00th=[  161],
     | 70.00th=[  161], 80.00th=[  161], 90.00th=[  180], 95.00th=[  200],
     | 99.00th=[  241], 99.50th=[  262], 99.90th=[  281], 99.95th=[  281],
     | 99.99th=[  281]
    lat (msec) : 2=0.03%, 20=0.63%, 50=1.00%, 100=1.00%, 250=96.62%
    lat (msec) : 500=0.73%
  cpu          : usr=0.01%, sys=0.03%, ctx=6047, majf=0, minf=290
  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    : total=r=3014/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=12056KB, aggrb=200KB/s, minb=200KB/s, maxb=200KB/s, mint=60160msec, maxt=60160msec

Disk stats (read/write):
  ploop46225p1: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=11122: Thu Aug  1 11:20:28 2019
  write: io=144828KB, bw=2413.6KB/s, iops=603, runt= 60007msec
    clat (usec): min=2, max=36999K, avg=10331.41, stdev=437479.93
     lat (usec): min=2, max=36999K, avg=10332.25, stdev=437479.97
    clat percentiles (usec):
     |  1.00th=[    3],  5.00th=[    4], 10.00th=[    5], 20.00th=[    5],
     | 30.00th=[    6], 40.00th=[    6], 50.00th=[    6], 60.00th=[    6],
     | 70.00th=[    7], 80.00th=[    9], 90.00th=[   16], 95.00th=[ 1992],
     | 99.00th=[100864], 99.50th=[140288], 99.90th=[1204224], 99.95th=[1548288],
     | 99.99th=[16711680]
    lat (usec) : 4=1.73%, 10=80.56%, 20=9.02%, 50=1.71%, 100=0.19%
    lat (usec) : 250=0.04%, 500=0.01%, 750=0.02%, 1000=0.25%
    lat (msec) : 2=1.50%, 4=1.66%, 10=1.55%, 20=0.40%, 50=0.10%
    lat (msec) : 100=0.22%, 250=0.90%, 500=0.01%, 750=0.01%, 1000=0.01%
    lat (msec) : 2000=0.09%, >=2000=0.04%
  cpu          : usr=0.02%, sys=0.11%, ctx=10006, majf=0, minf=286
  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    : total=r=0/w=36207/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=144828KB, aggrb=2413KB/s, minb=2413KB/s, maxb=2413KB/s, mint=60007msec, maxt=60007msec

Disk stats (read/write):
  ploop46225p1: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%

Speed​​test

Retrieving speedtest.net configuration...
Testing from NOTER-KLAV Private Enterprise (185.219.80.91)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by LinuxPro (Amsterdam) [1.13 km]: 7.442 ms
Testing download speed................................................................................
Download: 48.50 Mbit/s
Testing upload speed................................................................................................
Upload: 76.13 Mbit/s

Veesp の仮想サーバー:

安価な仮想サーバーのテスト

  • 機器はサンクトペテルブルクの独自の DC (オブホフスカヤ オボロナ通り) にあります。
  • HDD サンドボックス料金 (10 GB HDD / 1x 2.67 GHz CPU / 0.5 GB RAM / 200 Mbit/s、0.1 TB トラフィック、月額 65 ルーブル)。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-8-amd64 -- #1 SMP Debian 4.9.144-3.1 (2019-02-19)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Xeon(R) CPU X5650 @ 2.67GHz (5333.5 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   18:07:28 up 8 min,  1 user,  load average: 1.07, 1.00, 0.53; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Mon Sep 16 2019 18:07:28 - 18:35:37
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       25345643.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3438.1 MWIPS (9.9 s, 7 samples)
Execl Throughput                               3783.7 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        527507.6 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          151963.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1261205.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                              906688.5 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 116073.9 lps   (10.0 s, 7 samples)
Process Creation                               8518.9 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   6767.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    895.2 lpm   (60.1 s, 2 samples)
System Call Overhead                         648188.7 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   25345643.6   2171.9
Double-Precision Whetstone                       55.0       3438.1    625.1
Execl Throughput                                 43.0       3783.7    879.9
File Copy 1024 bufsize 2000 maxblocks          3960.0     527507.6   1332.1
File Copy 256 bufsize 500 maxblocks            1655.0     151963.5    918.2
File Copy 4096 bufsize 8000 maxblocks          5800.0    1261205.9   2174.5
Pipe Throughput                               12440.0     906688.5    728.8
Pipe-based Context Switching                   4000.0     116073.9    290.2
Process Creation                                126.0       8518.9    676.1
Shell Scripts (1 concurrent)                     42.4       6767.9   1596.2
Shell Scripts (8 concurrent)                      6.0        895.2   1492.0
System Call Overhead                          15000.0     648188.7    432.1
                                                                   ========
System Benchmarks Index Score                                         938.6

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 24.6856 s, 87.0 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 64.7868 s, 33.1 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=7870: Mon Sep 16 18:38:49 2019
  read : io=43940KB, bw=749484B/s, iops=182, runt= 60034msec
    clat (usec): min=59, max=394340, avg=43701.14, stdev=33411.67
     lat (usec): min=60, max=394341, avg=43702.00, stdev=33411.68
    clat percentiles (usec):
     |  1.00th=[  876],  5.00th=[ 7200], 10.00th=[11968], 20.00th=[20608],
     | 30.00th=[27776], 40.00th=[33024], 50.00th=[37632], 60.00th=[42240],
     | 70.00th=[48384], 80.00th=[58112], 90.00th=[77312], 95.00th=[103936],
     | 99.00th=[187392], 99.50th=[211968], 99.90th=[264192], 99.95th=[280576],
     | 99.99th=[325632]
    bw (KB  /s): min=   20, max=  208, per=12.56%, avg=91.80, stdev=23.94
    lat (usec) : 100=0.31%, 250=0.36%, 500=0.23%, 750=0.06%, 1000=0.06%
    lat (msec) : 2=0.26%, 4=1.20%, 10=5.49%, 20=11.40%, 50=52.59%
    lat (msec) : 100=22.58%, 250=5.28%, 500=0.17%
  cpu          : usr=0.05%, sys=0.07%, ctx=11013, majf=0, minf=75
  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    : total=r=10985/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=43940KB, aggrb=731KB/s, minb=731KB/s, maxb=731KB/s, mint=60034msec, maxt=60034msec

Disk stats (read/write):
  vda: ios=11133/8, merge=0/4, ticks=493100/428, in_queue=494484, util=96.94%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=7897: Mon Sep 16 18:42:30 2019
  write: io=133508KB, bw=1454.2KB/s, iops=363, runt= 91812msec
    clat (usec): min=2, max=40324K, avg=10794.49, stdev=460561.62
     lat (usec): min=2, max=40324K, avg=10795.16, stdev=460561.65
    clat percentiles (usec):
     |  1.00th=[    3],  5.00th=[    3], 10.00th=[    3], 20.00th=[    4],
     | 30.00th=[    4], 40.00th=[    4], 50.00th=[    4], 60.00th=[    4],
     | 70.00th=[    5], 80.00th=[    5], 90.00th=[    7], 95.00th=[   15],
     | 99.00th=[132096], 99.50th=[175104], 99.90th=[203776], 99.95th=[4685824],
     | 99.99th=[16711680]
    bw (KB  /s): min=    0, max=23414, per=100.00%, avg=1795.50, stdev=4287.44
    lat (usec) : 4=15.67%, 10=77.56%, 20=2.98%, 50=1.90%, 100=0.17%
    lat (usec) : 250=0.01%, 750=0.01%, 1000=0.01%
    lat (msec) : 2=0.01%, 4=0.01%, 10=0.07%, 20=0.28%, 50=0.18%
    lat (msec) : 100=0.06%, 250=1.00%, 500=0.02%, 1000=0.01%, 2000=0.01%
    lat (msec) : >=2000=0.06%
  cpu          : usr=0.02%, sys=0.13%, ctx=27584, majf=0, minf=83
  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    : total=r=0/w=33377/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=133508KB, aggrb=1454KB/s, minb=1454KB/s, maxb=1454KB/s, mint=91812msec, maxt=91812msec

Disk stats (read/write):
  vda: ios=169/15176, merge=0/12608, ticks=22232/115416, in_queue=137640, util=99.14%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=7883: Mon Sep 16 18:40:28 2019
  read : io=30852KB, bw=526198B/s, iops=128, runt= 60039msec
    clat (usec): min=66, max=686836, avg=62246.24, stdev=62045.44
     lat (usec): min=67, max=686837, avg=62247.16, stdev=62045.43
    clat percentiles (usec):
     |  1.00th=[ 1800],  5.00th=[ 6304], 10.00th=[10048], 20.00th=[18816],
     | 30.00th=[28288], 40.00th=[35584], 50.00th=[42752], 60.00th=[51456],
     | 70.00th=[67072], 80.00th=[94720], 90.00th=[140288], 95.00th=[185344],
     | 99.00th=[292864], 99.50th=[350208], 99.90th=[536576], 99.95th=[634880],
     | 99.99th=[684032]
    bw (KB  /s): min=    5, max=  163, per=12.67%, avg=65.01, stdev=31.07
    lat (usec) : 100=0.23%, 250=0.31%, 500=0.12%, 750=0.03%, 1000=0.06%
    lat (msec) : 2=0.38%, 4=1.66%, 10=7.07%, 20=11.20%, 50=37.83%
    lat (msec) : 100=22.47%, 250=17.01%, 500=1.53%, 750=0.10%
  cpu          : usr=0.02%, sys=0.06%, ctx=7738, majf=0, minf=83
  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    : total=r=7713/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=30852KB, aggrb=513KB/s, minb=513KB/s, maxb=513KB/s, mint=60039msec, maxt=60039msec

Disk stats (read/write):
  vda: ios=7669/8, merge=0/5, ticks=478044/828, in_queue=478972, util=99.88%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=7910: Mon Sep 16 18:43:33 2019
  write: io=32548KB, bw=555060B/s, iops=135, runt= 60046msec
    clat (usec): min=171, max=4236.3K, avg=59003.70, stdev=264488.45
     lat (usec): min=171, max=4236.3K, avg=59004.72, stdev=264488.53
    clat percentiles (usec):
     |  1.00th=[  262],  5.00th=[  358], 10.00th=[  474], 20.00th=[  860],
     | 30.00th=[ 1528], 40.00th=[ 3536], 50.00th=[ 5536], 60.00th=[ 8032],
     | 70.00th=[ 8896], 80.00th=[12864], 90.00th=[20864], 95.00th=[61184],
     | 99.00th=[1253376], 99.50th=[1712128], 99.90th=[2899968], 99.95th=[3325952],
     | 99.99th=[4227072]
    bw (KB  /s): min=    2, max=  229, per=14.10%, avg=76.41, stdev=32.24
    lat (usec) : 250=0.74%, 500=10.19%, 750=6.96%, 1000=4.38%
    lat (msec) : 2=11.12%, 4=7.95%, 10=32.91%, 20=15.37%, 50=4.98%
    lat (msec) : 100=0.73%, 250=0.11%, 500=0.07%, 750=0.61%, 1000=1.67%
    lat (msec) : 2000=1.86%, >=2000=0.36%
  cpu          : usr=0.03%, sys=0.09%, ctx=16695, majf=0, minf=76
  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    : total=r=0/w=8137/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=32548KB, aggrb=542KB/s, minb=542KB/s, maxb=542KB/s, mint=60046msec, maxt=60046msec

Disk stats (read/write):
  vda: ios=170/8180, merge=0/594, ticks=19048/60876, in_queue=79924, util=98.62%

Speed​​test

Retrieving speedtest.net configuration...
Testing from Fishnet Communications LLC (94.242.58.40)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Orange Business Services, Russia & CIS (Moscow) [1.61 km]: 12.636 ms
Testing download speed................................................................................
Download: 218.16 Mbit/s
Testing upload speed................................................................................................
Upload: 189.89 Mbit/s

PQ.Hosting の仮想サーバー:

安価な仮想サーバーのテスト

  • この機器はロシア (モスクワ、DC DataPro)、オランダ (メッペル、DC Serverius)、モルドバ (キシナウ、DC Alexhost)、ラトビア (リガ、DC VERnet) にあります。
  • アルミニウム料金 (15 GB SSD / 1x 2.8 GHz CPU / 1 GB RAM / 1000 Mbit/s、月額 1 ユーロ)。

テスト:

Unixベンチ (RU)

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-9-amd64 -- #1 SMP Debian 4.9.168-1+deb9u3 (2019-06-16)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version (cpu64-rhel6) (5586.5 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   23:15:17 up 11:28,  1 user,  load average: 0.00, 0.00, 0.00; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Mon Jun 24 2019 23:15:17 - 23:43:21
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       34123920.4 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     4320.5 MWIPS (9.9 s, 7 samples)
Execl Throughput                               4163.6 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        331033.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           92832.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1021086.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                              490283.2 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 117833.9 lps   (10.0 s, 7 samples)
Process Creation                              10061.3 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   7836.1 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    981.6 lpm   (60.0 s, 2 samples)
System Call Overhead                         326481.9 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   34123920.4   2924.1
Double-Precision Whetstone                       55.0       4320.5    785.5
Execl Throughput                                 43.0       4163.6    968.3
File Copy 1024 bufsize 2000 maxblocks          3960.0     331033.8    835.9
File Copy 256 bufsize 500 maxblocks            1655.0      92832.7    560.9
File Copy 4096 bufsize 8000 maxblocks          5800.0    1021086.9   1760.5
Pipe Throughput                               12440.0     490283.2    394.1
Pipe-based Context Switching                   4000.0     117833.9    294.6
Process Creation                                126.0      10061.3    798.5
Shell Scripts (1 concurrent)                     42.4       7836.1   1848.1
Shell Scripts (8 concurrent)                      6.0        981.6   1636.0
System Call Overhead                          15000.0     326481.9    217.7
                                                                   ========
System Benchmarks Index Score                                         833.2

Unixベンチ (NL)

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-9-amd64 -- #1 SMP Debian 4.9.168-1+deb9u3 (2019-06-16)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version (cpu64-rhel6) (4000.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   17:48:01 up 30 days, 19:48,  1 user,  load average: 0.79, 0.39, 0.16; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Sun Sep 01 2019 17:48:01 - 18:15:49
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       17944619.4 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2541.3 MWIPS (8.2 s, 7 samples)
Execl Throughput                               1955.3 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        152651.9 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           38882.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        445940.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                              224267.3 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  52138.8 lps   (10.0 s, 7 samples)
Process Creation                               4483.9 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   3753.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    486.1 lpm   (60.1 s, 2 samples)
System Call Overhead                         151653.5 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   17944619.4   1537.7
Double-Precision Whetstone                       55.0       2541.3    462.1
Execl Throughput                                 43.0       1955.3    454.7
File Copy 1024 bufsize 2000 maxblocks          3960.0     152651.9    385.5
File Copy 256 bufsize 500 maxblocks            1655.0      38882.7    234.9
File Copy 4096 bufsize 8000 maxblocks          5800.0     445940.9    768.9
Pipe Throughput                               12440.0     224267.3    180.3
Pipe-based Context Switching                   4000.0      52138.8    130.3
Process Creation                                126.0       4483.9    355.9
Shell Scripts (1 concurrent)                     42.4       3753.9    885.4
Shell Scripts (8 concurrent)                      6.0        486.1    810.2
System Call Overhead                          15000.0     151653.5    101.1
                                                                   ========
System Benchmarks Index Score                                         393.2

Unixベンチ (MD)

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-9-amd64 -- #1 SMP Debian 4.9.168-1+deb9u3 (2019-06-16)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version (cpu64-rhel6) (5600.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   17:51:08 up 13 days,  3:34,  1 user,  load average: 1.07, 0.68, 0.33; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Sun Sep 01 2019 17:51:08 - 18:19:23
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       22940127.9 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3493.9 MWIPS (9.8 s, 7 samples)
Execl Throughput                               2580.9 lps   (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        190422.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           52531.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        624248.2 KBps  (30.0 s, 2 samples)
Pipe Throughput                              319487.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  76614.0 lps   (10.0 s, 7 samples)
Process Creation                               5971.0 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4097.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    534.9 lpm   (60.0 s, 2 samples)
System Call Overhead                         215310.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   22940127.9   1965.7
Double-Precision Whetstone                       55.0       3493.9    635.3
Execl Throughput                                 43.0       2580.9    600.2
File Copy 1024 bufsize 2000 maxblocks          3960.0     190422.3    480.9
File Copy 256 bufsize 500 maxblocks            1655.0      52531.5    317.4
File Copy 4096 bufsize 8000 maxblocks          5800.0     624248.2   1076.3
Pipe Throughput                               12440.0     319487.7    256.8
Pipe-based Context Switching                   4000.0      76614.0    191.5
Process Creation                                126.0       5971.0    473.9
Shell Scripts (1 concurrent)                     42.4       4097.2    966.3
Shell Scripts (8 concurrent)                      6.0        534.9    891.5
System Call Overhead                          15000.0     215310.8    143.5
                                                                   ========
System Benchmarks Index Score                                         515.8

Unixベンチ(LV)

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-11-amd64 -- #1 SMP Debian 4.9.189-3 (2019-09-02)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version (cpu64-rhel6) (5586.5 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   18:46:26 up 4 days, 21:45,  1 user,  load average: 0.53, 0.14, 0.05; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Mon Sep 16 2019 18:46:26 - 19:14:30
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       34241148.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     4298.3 MWIPS (9.9 s, 7 samples)
Execl Throughput                               3739.1 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        293843.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           80365.6 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        885168.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              404089.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 109078.4 lps   (10.0 s, 7 samples)
Process Creation                               9123.2 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   6840.3 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    889.1 lpm   (60.0 s, 2 samples)
System Call Overhead                         249437.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   34241148.6   2934.1
Double-Precision Whetstone                       55.0       4298.3    781.5
Execl Throughput                                 43.0       3739.1    869.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     293843.0    742.0
File Copy 256 bufsize 500 maxblocks            1655.0      80365.6    485.6
File Copy 4096 bufsize 8000 maxblocks          5800.0     885168.5   1526.2
Pipe Throughput                               12440.0     404089.8    324.8
Pipe-based Context Switching                   4000.0     109078.4    272.7
Process Creation                                126.0       9123.2    724.1
Shell Scripts (1 concurrent)                     42.4       6840.3   1613.3
Shell Scripts (8 concurrent)                      6.0        889.1   1481.8
System Call Overhead                          15000.0     249437.8    166.3
                                                                   ========
System Benchmarks Index Score                                         742.2

dd (RU)1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 5.20811 s, 412 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 5.7549 s, 373 MB/s

dd (オランダ語)1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 14.9915 s, 143 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 14.8393 s, 145 MB/s

DD(MD)1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 8.36403 s, 257 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 19.328 s, 111 MB/s

dd(LV)1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 5.94689 s, 361 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 7.10417 s, 302 MB/s

フィオ (RU)読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=20549: Mon Jun 24 23:44:39 2019
  read : io=8230.4MB, bw=140461KB/s, iops=35115, runt= 60001msec
    clat (usec): min=3, max=13440, avg=220.09, stdev=168.87
     lat (usec): min=4, max=21543, avg=221.70, stdev=169.80
    clat percentiles (usec):
     |  1.00th=[    4],  5.00th=[    6], 10.00th=[  113], 20.00th=[  149],
     | 30.00th=[  169], 40.00th=[  187], 50.00th=[  205], 60.00th=[  223],
     | 70.00th=[  249], 80.00th=[  282], 90.00th=[  334], 95.00th=[  386],
     | 99.00th=[  564], 99.50th=[  716], 99.90th=[ 2736], 99.95th=[ 3216],
     | 99.99th=[ 4256]
    lat (usec) : 4=0.13%, 10=5.20%, 20=0.85%, 50=0.10%, 100=1.87%
    lat (usec) : 250=62.54%, 500=27.78%, 750=1.09%, 1000=0.15%
    lat (msec) : 2=0.10%, 4=0.20%, 10=0.01%, 20=0.01%
  cpu          : usr=3.46%, sys=8.52%, ctx=1981547, majf=3, minf=95
  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    : total=r=2106957/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=8230.4MB, aggrb=140461KB/s, minb=140461KB/s, maxb=140461KB/s, mint=60001msec, maxt=60001msec

Disk stats (read/write):
  vda: ios=1974597/13, merge=0/3, ticks=96360/0, in_queue=95732, util=90.63%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=20575: Mon Jun 24 23:46:40 2019
  write: io=18781MB, bw=319092KB/s, iops=79772, runt= 60269msec
    clat (usec): min=3, max=351032, avg=62.03, stdev=1149.91
     lat (usec): min=4, max=360051, avg=70.29, stdev=1224.93
    clat percentiles (usec):
     |  1.00th=[    4],  5.00th=[    4], 10.00th=[    4], 20.00th=[    4],
     | 30.00th=[    4], 40.00th=[    5], 50.00th=[    5], 60.00th=[    5],
     | 70.00th=[    5], 80.00th=[    5], 90.00th=[    6], 95.00th=[    7],
     | 99.00th=[   13], 99.50th=[ 1272], 99.90th=[16512], 99.95th=[21376],
     | 99.99th=[30080]
    lat (usec) : 4=0.57%, 10=97.74%, 20=0.89%, 50=0.08%, 100=0.08%
    lat (usec) : 250=0.05%, 500=0.05%, 750=0.02%, 1000=0.01%
    lat (msec) : 2=0.04%, 4=0.07%, 10=0.17%, 20=0.17%, 50=0.06%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%
  cpu          : usr=4.92%, sys=5.80%, ctx=85278, majf=0, minf=54
  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    : total=r=0/w=4807836/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=18781MB, aggrb=319091KB/s, minb=319091KB/s, maxb=319091KB/s, mint=60269msec, maxt=60269msec

Disk stats (read/write):
  vda: ios=363/1745565, merge=0/10995, ticks=16/4723708, in_queue=4752280, util=70.07%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes

benchmark: (groupid=0, jobs=8): err= 0: pid=20560: Mon Jun 24 23:45:39 2019
  read : io=8760.7MB, bw=149512KB/s, iops=37378, runt= 60001msec
    clat (usec): min=29, max=20597, avg=205.81, stdev=110.27
     lat (usec): min=30, max=20599, avg=207.40, stdev=110.93
    clat percentiles (usec):
     |  1.00th=[   82],  5.00th=[  110], 10.00th=[  126], 20.00th=[  147],
     | 30.00th=[  163], 40.00th=[  175], 50.00th=[  189], 60.00th=[  203],
     | 70.00th=[  225], 80.00th=[  253], 90.00th=[  298], 95.00th=[  338],
     | 99.00th=[  494], 99.50th=[  588], 99.90th=[ 1080], 99.95th=[ 1736],
     | 99.99th=[ 3984]
    lat (usec) : 50=0.02%, 100=2.80%, 250=76.22%, 500=20.00%, 750=0.73%
    lat (usec) : 1000=0.11%
    lat (msec) : 2=0.07%, 4=0.03%, 10=0.01%, 20=0.01%, 50=0.01%
  cpu          : usr=3.73%, sys=8.41%, ctx=2248153, majf=0, minf=58
  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    : total=r=2242722/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=8760.7MB, aggrb=149512KB/s, minb=149512KB/s, maxb=149512KB/s, mint=60001msec, maxt=60001msec

Disk stats (read/write):
  vda: ios=2237967/9, merge=0/8, ticks=105964/0, in_queue=105332, util=93.73%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes

benchmark: (groupid=0, jobs=8): err= 0: pid=20592: Mon Jun 24 23:47:41 2019
  write: io=2217.5MB, bw=37841KB/s, iops=9460, runt= 60004msec
    clat (usec): min=55, max=280991, avg=834.96, stdev=5125.29
     lat (usec): min=56, max=280992, avg=836.46, stdev=5125.35
    clat percentiles (usec):
     |  1.00th=[   64],  5.00th=[   70], 10.00th=[   74], 20.00th=[   79],
     | 30.00th=[   83], 40.00th=[   84], 50.00th=[   86], 60.00th=[   89],
     | 70.00th=[   93], 80.00th=[  102], 90.00th=[  123], 95.00th=[  167],
     | 99.00th=[26240], 99.50th=[27264], 99.90th=[30080], 99.95th=[31872],
     | 99.99th=[230400]
    lat (usec) : 100=78.04%, 250=18.60%, 500=0.47%, 750=0.04%, 1000=0.03%
    lat (msec) : 2=0.05%, 4=0.01%, 10=0.01%, 20=0.01%, 50=2.73%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%
  cpu          : usr=0.98%, sys=3.50%, ctx=1151513, majf=0, minf=68
  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    : total=r=0/w=567657/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=2217.5MB, aggrb=37841KB/s, minb=37841KB/s, maxb=37841KB/s, mint=60004msec, maxt=60004msec

Disk stats (read/write):
  vda: ios=0/575378, merge=0/20, ticks=0/64376, in_queue=64272, util=72.92%

フィオ (オランダ)読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=31842: Sun Sep  1 18:17:37 2019
  read : io=2425.7MB, bw=41395KB/s, iops=10348, runt= 60003msec
    clat (usec): min=5, max=825845, avg=735.78, stdev=3960.80
     lat (usec): min=7, max=825848, avg=741.91, stdev=3967.13
    clat percentiles (usec):
     |  1.00th=[    8],  5.00th=[  147], 10.00th=[  221], 20.00th=[  298],
     | 30.00th=[  350], 40.00th=[  410], 50.00th=[  478], 60.00th=[  564],
     | 70.00th=[  676], 80.00th=[  852], 90.00th=[ 1192], 95.00th=[ 1704],
     | 99.00th=[ 4384], 99.50th=[ 6688], 99.90th=[19584], 99.95th=[29056],
     | 99.99th=[90624]
    bw (KB  /s): min=   35, max= 8400, per=12.73%, avg=5269.91, stdev=2141.82
    lat (usec) : 10=2.32%, 20=1.44%, 50=0.16%, 100=0.10%, 250=9.09%
    lat (usec) : 500=39.62%, 750=21.80%, 1000=11.07%
    lat (msec) : 2=10.71%, 4=2.52%, 10=0.90%, 20=0.17%, 50=0.07%
    lat (msec) : 100=0.02%, 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
  cpu          : usr=3.08%, sys=6.95%, ctx=592221, majf=0, minf=76
  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    : total=r=620961/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=2425.7MB, aggrb=41395KB/s, minb=41395KB/s, maxb=41395KB/s, mint=60003msec, maxt=60003msec

Disk stats (read/write):
  vda: ios=594911/18, merge=0/14, ticks=171728/52, in_queue=171344, util=84.01%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=31871: Sun Sep  1 18:19:59 2019
  write: io=5841.6MB, bw=99669KB/s, iops=24917, runt= 60016msec
    clat (usec): min=5, max=1182.1K, avg=185.99, stdev=4524.79
     lat (usec): min=7, max=1182.1K, avg=208.32, stdev=4665.58
    clat percentiles (usec):
     |  1.00th=[    7],  5.00th=[    7], 10.00th=[    8], 20.00th=[    9],
     | 30.00th=[   10], 40.00th=[   11], 50.00th=[   11], 60.00th=[   11],
     | 70.00th=[   12], 80.00th=[   12], 90.00th=[   14], 95.00th=[   17],
     | 99.00th=[ 1384], 99.50th=[12096], 99.90th=[28032], 99.95th=[33536],
     | 99.99th=[105984]
    bw (KB  /s): min=    0, max=21832, per=13.67%, avg=13624.13, stdev=4833.63
    lat (usec) : 10=21.53%, 20=74.46%, 50=2.26%, 100=0.30%, 250=0.30%
    lat (usec) : 500=0.06%, 750=0.03%, 1000=0.02%
    lat (msec) : 2=0.06%, 4=0.10%, 10=0.28%, 20=0.35%, 50=0.22%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%, 750=0.01%, 2000=0.01%
  cpu          : usr=4.67%, sys=5.46%, ctx=46646, majf=0, minf=69
  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    : total=r=0/w=1495438/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=5841.6MB, aggrb=99669KB/s, minb=99669KB/s, maxb=99669KB/s, mint=60016msec, maxt=60016msec

Disk stats (read/write):
  vda: ios=293/392197, merge=0/6829, ticks=1880/4791112, in_queue=4793768, util=82.90%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=31858: Sun Sep  1 18:18:59 2019
  read : io=2344.4MB, bw=40008KB/s, iops=10001, runt= 60004msec
    clat (usec): min=53, max=1300.2K, avg=763.80, stdev=3474.39
     lat (usec): min=55, max=1300.2K, avg=769.36, stdev=3476.61
    clat percentiles (usec):
     |  1.00th=[  143],  5.00th=[  207], 10.00th=[  253], 20.00th=[  318],
     | 30.00th=[  378], 40.00th=[  450], 50.00th=[  540], 60.00th=[  652],
     | 70.00th=[  772], 80.00th=[  940], 90.00th=[ 1288], 95.00th=[ 1784],
     | 99.00th=[ 3856], 99.50th=[ 5600], 99.90th=[13120], 99.95th=[19328],
     | 99.99th=[44800]
    bw (KB  /s): min=    9, max= 8496, per=12.69%, avg=5076.46, stdev=1793.14
    lat (usec) : 100=0.10%, 250=9.49%, 500=35.90%, 750=22.95%, 1000=14.00%
    lat (msec) : 2=13.76%, 4=2.87%, 10=0.75%, 20=0.13%, 50=0.04%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
    lat (msec) : 2000=0.01%
  cpu          : usr=3.23%, sys=6.54%, ctx=601201, majf=0, minf=76
  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    : total=r=600154/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=2344.4MB, aggrb=40007KB/s, minb=40007KB/s, maxb=40007KB/s, mint=60004msec, maxt=60004msec

Disk stats (read/write):
  vda: ios=598966/9, merge=0/5, ticks=194768/4, in_queue=194352, util=86.55%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=31896: Sun Sep  1 18:21:01 2019
  write: io=758060KB, bw=12632KB/s, iops=3158, runt= 60011msec
    clat (usec): min=98, max=533622, avg=2511.65, stdev=11053.52
     lat (usec): min=100, max=533625, avg=2515.25, stdev=11054.15
    clat percentiles (usec):
     |  1.00th=[  131],  5.00th=[  143], 10.00th=[  155], 20.00th=[  177],
     | 30.00th=[  197], 40.00th=[  217], 50.00th=[  237], 60.00th=[  266],
     | 70.00th=[  302], 80.00th=[  370], 90.00th=[  644], 95.00th=[25728],
     | 99.00th=[32384], 99.50th=[35584], 99.90th=[48896], 99.95th=[119296],
     | 99.99th=[460800]
    bw (KB  /s): min=  106, max= 2561, per=12.73%, avg=1608.20, stdev=391.86
    lat (usec) : 100=0.01%, 250=54.90%, 500=32.31%, 750=3.63%, 1000=0.85%
    lat (msec) : 2=0.59%, 4=0.16%, 10=0.09%, 20=0.10%, 50=7.27%
    lat (msec) : 100=0.04%, 250=0.01%, 500=0.03%, 750=0.01%
  cpu          : usr=0.84%, sys=3.27%, ctx=394058, majf=0, minf=67
  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    : total=r=0/w=189515/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=758060KB, aggrb=12632KB/s, minb=12632KB/s, maxb=12632KB/s, mint=60011msec, maxt=60011msec

Disk stats (read/write):
  vda: ios=135/189261, merge=0/2310, ticks=260/47292, in_queue=47780, util=72.85%

フィオ (MD)読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=725: Sun Sep  1 18:21:06 2019
  read : io=3620.1MB, bw=61795KB/s, iops=15448, runt= 60002msec
    clat (usec): min=4, max=421928, avg=499.00, stdev=1567.53
     lat (usec): min=5, max=421929, avg=502.28, stdev=1570.05
    clat percentiles (usec):
     |  1.00th=[    6],  5.00th=[  120], 10.00th=[  177], 20.00th=[  235],
     | 30.00th=[  274], 40.00th=[  314], 50.00th=[  358], 60.00th=[  410],
     | 70.00th=[  486], 80.00th=[  596], 90.00th=[  836], 95.00th=[ 1128],
     | 99.00th=[ 2832], 99.50th=[ 4320], 99.90th=[ 8768], 99.95th=[11968],
     | 99.99th=[34048]
    bw (KB  /s): min=   63, max=11864, per=12.53%, avg=7743.07, stdev=2025.73
    lat (usec) : 10=2.82%, 20=0.24%, 50=0.21%, 100=0.65%, 250=19.78%
    lat (usec) : 500=48.02%, 750=15.78%, 1000=5.98%
    lat (msec) : 2=4.82%, 4=1.12%, 10=0.51%, 20=0.06%, 50=0.01%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%
  cpu          : usr=3.16%, sys=6.70%, ctx=895140, majf=0, minf=76
  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    : total=r=926956/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=3620.1MB, aggrb=61795KB/s, minb=61795KB/s, maxb=61795KB/s, mint=60002msec, maxt=60002msec

Disk stats (read/write):
  vda: ios=895284/9, merge=0/6, ticks=192912/48, in_queue=192488, util=88.29%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=751: Sun Sep  1 18:23:20 2019
  write: io=5957.5MB, bw=101655KB/s, iops=25413, runt= 60011msec
    clat (usec): min=4, max=3615.9K, avg=210.69, stdev=12600.90
     lat (usec): min=5, max=3615.9K, avg=228.05, stdev=12685.41
    clat percentiles (usec):
     |  1.00th=[    5],  5.00th=[    5], 10.00th=[    6], 20.00th=[    7],
     | 30.00th=[    7], 40.00th=[    8], 50.00th=[    8], 60.00th=[    8],
     | 70.00th=[    9], 80.00th=[    9], 90.00th=[   10], 95.00th=[   13],
     | 99.00th=[  147], 99.50th=[ 7968], 99.90th=[28032], 99.95th=[36096],
     | 99.99th=[254976]
    bw (KB  /s): min=    0, max=28111, per=15.41%, avg=15665.93, stdev=7515.70
    lat (usec) : 10=84.10%, 20=13.19%, 50=1.12%, 100=0.42%, 250=0.27%
    lat (usec) : 500=0.07%, 750=0.03%, 1000=0.02%
    lat (msec) : 2=0.06%, 4=0.09%, 10=0.21%, 20=0.23%, 50=0.17%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
    lat (msec) : 2000=0.01%, >=2000=0.01%
  cpu          : usr=3.78%, sys=4.40%, ctx=44638, majf=0, minf=70
  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    : total=r=0/w=1525105/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=5957.5MB, aggrb=101655KB/s, minb=101655KB/s, maxb=101655KB/s, mint=60011msec, maxt=60011msec

Disk stats (read/write):
  vda: ios=260/271490, merge=0/9232, ticks=3700/6093556, in_queue=6097168, util=94.72%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=738: Sun Sep  1 18:22:19 2019
  read : io=3310.5MB, bw=56495KB/s, iops=14123, runt= 60003msec
    clat (usec): min=41, max=56402, avg=545.26, stdev=795.74
     lat (usec): min=42, max=56404, avg=548.75, stdev=801.70
    clat percentiles (usec):
     |  1.00th=[  104],  5.00th=[  151], 10.00th=[  187], 20.00th=[  239],
     | 30.00th=[  282], 40.00th=[  326], 50.00th=[  382], 60.00th=[  454],
     | 70.00th=[  540], 80.00th=[  668], 90.00th=[  924], 95.00th=[ 1272],
     | 99.00th=[ 3184], 99.50th=[ 4832], 99.90th=[10304], 99.95th=[13888],
     | 99.99th=[24192]
    bw (KB  /s): min= 3608, max=10648, per=12.52%, avg=7072.14, stdev=1259.64
    lat (usec) : 50=0.01%, 100=0.75%, 250=21.84%, 500=42.72%, 750=18.85%
    lat (usec) : 1000=7.58%
    lat (msec) : 2=6.04%, 4=1.54%, 10=0.57%, 20=0.09%, 50=0.02%
    lat (msec) : 100=0.01%
  cpu          : usr=3.02%, sys=6.29%, ctx=845766, majf=0, minf=74
  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    : total=r=847462/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=3310.5MB, aggrb=56494KB/s, minb=56494KB/s, maxb=56494KB/s, mint=60003msec, maxt=60003msec

Disk stats (read/write):
  vda: ios=846682/11, merge=0/6, ticks=224952/40, in_queue=224480, util=89.22%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=768: Sun Sep  1 18:24:21 2019
  write: io=912340KB, bw=15195KB/s, iops=3798, runt= 60043msec
    clat (usec): min=70, max=2025.8K, avg=2091.37, stdev=19475.86
     lat (usec): min=72, max=2025.8K, avg=2093.79, stdev=19475.96
    clat percentiles (usec):
     |  1.00th=[   93],  5.00th=[  104], 10.00th=[  113], 20.00th=[  129],
     | 30.00th=[  141], 40.00th=[  153], 50.00th=[  165], 60.00th=[  183],
     | 70.00th=[  207], 80.00th=[  253], 90.00th=[  418], 95.00th=[22400],
     | 99.00th=[35072], 99.50th=[38656], 99.90th=[48384], 99.95th=[79360],
     | 99.99th=[1011712]
    bw (KB  /s): min=   33, max= 3175, per=13.41%, avg=2037.35, stdev=545.73
    lat (usec) : 100=2.70%, 250=76.85%, 500=11.94%, 750=1.65%, 1000=0.54%
    lat (msec) : 2=0.56%, 4=0.30%, 10=0.15%, 20=0.04%, 50=5.17%
    lat (msec) : 100=0.05%, 250=0.01%, 500=0.02%, 750=0.01%, 2000=0.01%
    lat (msec) : >=2000=0.01%
  cpu          : usr=0.71%, sys=2.70%, ctx=468028, majf=0, minf=84
  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    : total=r=0/w=228085/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=912340KB, aggrb=15194KB/s, minb=15194KB/s, maxb=15194KB/s, mint=60043msec, maxt=60043msec

Disk stats (read/write):
  vda: ios=251/228180, merge=0/2274, ticks=54132/54884, in_queue=109212, util=77.85%

フィオ(LV)読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=30260: Mon Sep 16 19:15:51 2019
  read : io=7298.2MB, bw=124552KB/s, iops=31138, runt= 60001msec
    clat (usec): min=3, max=9331, avg=246.19, stdev=184.71
     lat (usec): min=4, max=9332, avg=248.14, stdev=185.21
    clat percentiles (usec):
     |  1.00th=[    4],  5.00th=[    5], 10.00th=[  118], 20.00th=[  163],
     | 30.00th=[  187], 40.00th=[  209], 50.00th=[  229], 60.00th=[  249],
     | 70.00th=[  274], 80.00th=[  318], 90.00th=[  386], 95.00th=[  450],
     | 99.00th=[  700], 99.50th=[ 1020], 99.90th=[ 2768], 99.95th=[ 3088],
     | 99.99th=[ 4320]
    bw (KB  /s): min= 9640, max=19312, per=12.50%, avg=15564.41, stdev=1923.50
    lat (usec) : 4=0.03%, 10=6.21%, 20=1.06%, 50=0.19%, 100=1.00%
    lat (usec) : 250=51.87%, 500=36.50%, 750=2.29%, 1000=0.35%
    lat (msec) : 2=0.29%, 4=0.21%, 10=0.02%
  cpu          : usr=4.11%, sys=7.96%, ctx=1736633, majf=2, minf=79
  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    : total=r=1868316/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=7298.2MB, aggrb=124552KB/s, minb=124552KB/s, maxb=124552KB/s, mint=60001msec, maxt=60001msec

Disk stats (read/write):
  vda: ios=1723192/7, merge=0/6, ticks=92616/0, in_queue=92060, util=89.49%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=30298: Mon Sep 16 19:17:59 2019
  write: io=15828MB, bw=270089KB/s, iops=67522, runt= 60009msec
    clat (usec): min=3, max=422856, avg=67.39, stdev=1253.64
     lat (usec): min=4, max=422858, avg=76.81, stdev=1334.49
    clat percentiles (usec):
     |  1.00th=[    4],  5.00th=[    4], 10.00th=[    4], 20.00th=[    4],
     | 30.00th=[    5], 40.00th=[    5], 50.00th=[    5], 60.00th=[    5],
     | 70.00th=[    5], 80.00th=[    6], 90.00th=[    7], 95.00th=[    9],
     | 99.00th=[   21], 99.50th=[ 1752], 99.90th=[18304], 99.95th=[23936],
     | 99.99th=[32128]
    bw (KB  /s): min=    0, max=45536, per=12.74%, avg=34396.28, stdev=6030.11
    lat (usec) : 4=0.06%, 10=96.62%, 20=2.29%, 50=0.26%, 100=0.09%
    lat (usec) : 250=0.06%, 500=0.05%, 750=0.03%, 1000=0.02%
    lat (msec) : 2=0.05%, 4=0.07%, 10=0.16%, 20=0.17%, 50=0.08%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%
  cpu          : usr=5.44%, sys=5.56%, ctx=82279, majf=0, minf=84
  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    : total=r=0/w=4051946/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=15828MB, aggrb=270089KB/s, minb=270089KB/s, maxb=270089KB/s, mint=60009msec, maxt=60009msec

Disk stats (read/write):
  vda: ios=563/1293995, merge=0/12052, ticks=1668/3813796, in_queue=3815344, util=58.45%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=30277: Mon Sep 16 19:16:58 2019
  read : io=7164.9MB, bw=122276KB/s, iops=30569, runt= 60002msec
    clat (usec): min=32, max=26274, avg=250.65, stdev=211.58
     lat (usec): min=33, max=26276, avg=252.67, stdev=212.56
    clat percentiles (usec):
     |  1.00th=[   91],  5.00th=[  122], 10.00th=[  141], 20.00th=[  165],
     | 30.00th=[  185], 40.00th=[  203], 50.00th=[  219], 60.00th=[  239],
     | 70.00th=[  266], 80.00th=[  306], 90.00th=[  374], 95.00th=[  438],
     | 99.00th=[  740], 99.50th=[ 1048], 99.90th=[ 2672], 99.95th=[ 3856],
     | 99.99th=[ 7392]
    bw (KB  /s): min= 3904, max=19992, per=12.50%, avg=15288.18, stdev=2617.92
    lat (usec) : 50=0.02%, 100=1.70%, 250=62.96%, 500=32.36%, 750=1.99%
    lat (usec) : 1000=0.43%
    lat (msec) : 2=0.37%, 4=0.12%, 10=0.04%, 20=0.01%, 50=0.01%
  cpu          : usr=4.16%, sys=7.65%, ctx=1839930, majf=0, minf=74
  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    : total=r=1834207/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=7164.9MB, aggrb=122276KB/s, minb=122276KB/s, maxb=122276KB/s, mint=60002msec, maxt=60002msec

Disk stats (read/write):
  vda: ios=1831922/10, merge=0/5, ticks=111128/0, in_queue=110548, util=92.60%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=30314: Mon Sep 16 19:18:59 2019
  write: io=2170.5MB, bw=37039KB/s, iops=9259, runt= 60005msec
    clat (usec): min=55, max=281711, avg=854.04, stdev=4932.61
     lat (usec): min=56, max=281713, avg=855.74, stdev=4932.69
    clat percentiles (usec):
     |  1.00th=[   63],  5.00th=[   68], 10.00th=[   71], 20.00th=[   79],
     | 30.00th=[   84], 40.00th=[   85], 50.00th=[   87], 60.00th=[   90],
     | 70.00th=[   95], 80.00th=[  105], 90.00th=[  125], 95.00th=[  181],
     | 99.00th=[24704], 99.50th=[25472], 99.90th=[27776], 99.95th=[29568],
     | 99.99th=[201728]
    bw (KB  /s): min= 1988, max= 6319, per=12.57%, avg=4657.35, stdev=751.24
    lat (usec) : 100=74.81%, 250=21.30%, 500=0.65%, 750=0.06%, 1000=0.02%
    lat (msec) : 2=0.03%, 4=0.01%, 10=0.01%, 20=0.26%, 50=2.84%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%
  cpu          : usr=1.21%, sys=3.69%, ctx=1128730, majf=0, minf=75
  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    : total=r=0/w=555638/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=2170.5MB, aggrb=37039KB/s, minb=37039KB/s, maxb=37039KB/s, mint=60005msec, maxt=60005msec

Disk stats (read/write):
  vda: ios=201/555186, merge=0/1670, ticks=264/43420, in_queue=43564, util=68.46%

スピードテスト (RU)

Retrieving speedtest.net configuration...
Testing from Webhost LLC (45.67.230.12)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Moldtelecom JSC (Chisinau) [8.15 km]: 34.064 ms
Testing download speed................................................................................
Download: 65.92 Mbit/s
Testing upload speed................................................................................................
Upload: 74.83 Mbit/s

スピードテスト (NL)

Retrieving speedtest.net configuration...
Testing from Pq Hosting S.r.l. (45.67.228.18)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Solcon (Dronten) [37.27 km]: 24.408 ms
Testing download speed................................................................................
Download: 532.35 Mbit/s
Testing upload speed................................................................................................
Upload: 243.09 Mbit/s

スピードテスト (MD)

Retrieving speedtest.net configuration...
Testing from Pq Hosting S.r.l. (45.67.229.16)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by trabia network (Chisinau) [4.54 km]: 27.908 ms
Testing download speed................................................................................
Download: 576.51 Mbit/s
Testing upload speed................................................................................................
Upload: 448.82 Mbit/s

スピードテスト(LV)

Retrieving speedtest.net configuration...
Testing from Pq Hosting S.r.l. (45.142.213.19)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by AS Balticom (Riga) [54.59 km]: 21.875 ms
Testing download speed................................................................................
Download: 159.01 Mbit/s
Testing upload speed................................................................................................
Upload: 177.19 Mbit/s

Good-Host の仮想サーバー:

安価な仮想サーバーのテスト

  • フランスの設備 (Gravlines、DC OVH)。
  • FR-KVM-Micro 料金表 (5 GB SSD / 1x 4.2 GHz CPU / 0.4 GB RAM / 50 Mbit/s で月額 79 ルーブル)。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-8-amd64 -- #1 SMP Debian 4.9.144-3 (2019-02-02)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version 1.5.3 (8400.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   17:16:59 up 95 days,  5:52,  1 user,  load average: 0.75, 0.27, 0.10; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Thu Aug 08 2019 17:16:59 - 17:46:19
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       44766634.3 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3458.0 MWIPS (16.7 s, 7 samples)
Execl Throughput                               4181.1 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        327836.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           90502.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1034806.4 KBps  (30.0 s, 2 samples)
Pipe Throughput                              457766.5 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 101717.1 lps   (10.0 s, 7 samples)
Process Creation                              10014.0 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   7089.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    857.8 lpm   (60.0 s, 2 samples)
System Call Overhead                         437065.5 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   44766634.3   3836.0
Double-Precision Whetstone                       55.0       3458.0    628.7
Execl Throughput                                 43.0       4181.1    972.4
File Copy 1024 bufsize 2000 maxblocks          3960.0     327836.3    827.9
File Copy 256 bufsize 500 maxblocks            1655.0      90502.7    546.8
File Copy 4096 bufsize 8000 maxblocks          5800.0    1034806.4   1784.1
Pipe Throughput                               12440.0     457766.5    368.0
Pipe-based Context Switching                   4000.0     101717.1    254.3
Process Creation                                126.0      10014.0    794.8
Shell Scripts (1 concurrent)                     42.4       7089.9   1672.1
Shell Scripts (8 concurrent)                      6.0        857.8   1429.6
System Call Overhead                          15000.0     437065.5    291.4
                                                                   ========
System Benchmarks Index Score                                         824.1

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 3.95631 s, 543 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 5.13958 s, 418 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=28344: Thu Aug  8 17:47:34 2019
  read : io=3617.8MB, bw=61740KB/s, iops=15434, runt= 60003msec
    clat (usec): min=2, max=20244, avg=507.39, stdev=657.11
     lat (usec): min=2, max=20252, avg=509.08, stdev=658.24
    clat percentiles (usec):
     |  1.00th=[   43],  5.00th=[   88], 10.00th=[  119], 20.00th=[  183],
     | 30.00th=[  235], 40.00th=[  270], 50.00th=[  318], 60.00th=[  390],
     | 70.00th=[  502], 80.00th=[  700], 90.00th=[ 1048], 95.00th=[ 1384],
     | 99.00th=[ 3088], 99.50th=[ 4192], 99.90th=[ 8640], 99.95th=[ 9792],
     | 99.99th=[13120]
    bw (KB  /s): min= 2936, max=12160, per=12.52%, avg=7729.75, stdev=1794.96
    lat (usec) : 4=0.32%, 10=0.51%, 20=0.04%, 50=0.35%, 100=5.57%
    lat (usec) : 250=27.40%, 500=35.58%, 750=12.24%, 1000=7.03%
    lat (msec) : 2=8.72%, 4=1.67%, 10=0.52%, 20=0.04%, 50=0.01%
  cpu          : usr=1.78%, sys=5.37%, ctx=928767, majf=0, minf=74
  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    : total=r=926143/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=3617.8MB, aggrb=61739KB/s, minb=61739KB/s, maxb=61739KB/s, mint=60003msec, maxt=60003msec

Disk stats (read/write):
  vda: ios=918065/11, merge=0/8, ticks=298336/0, in_queue=298004, util=85.51%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=28370: Thu Aug  8 17:49:40 2019
  write: io=14871MB, bw=253765KB/s, iops=63441, runt= 60007msec
    clat (usec): min=2, max=340674, avg=85.73, stdev=1560.02
     lat (usec): min=3, max=340675, avg=92.59, stdev=1615.60
    clat percentiles (usec):
     |  1.00th=[    3],  5.00th=[    3], 10.00th=[    3], 20.00th=[    4],
     | 30.00th=[    4], 40.00th=[    4], 50.00th=[    4], 60.00th=[    5],
     | 70.00th=[    5], 80.00th=[    6], 90.00th=[    7], 95.00th=[    8],
     | 99.00th=[  101], 99.50th=[ 4832], 99.90th=[17536], 99.95th=[24448],
     | 99.99th=[62208]
    bw (KB  /s): min=    0, max=44688, per=12.85%, avg=32599.96, stdev=9507.40
    lat (usec) : 4=14.30%, 10=82.52%, 20=1.89%, 50=0.16%, 100=0.12%
    lat (usec) : 250=0.13%, 500=0.10%, 750=0.04%, 1000=0.02%
    lat (msec) : 2=0.06%, 4=0.11%, 10=0.28%, 20=0.19%, 50=0.07%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%
  cpu          : usr=3.01%, sys=6.04%, ctx=93095, majf=0, minf=68
  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    : total=r=0/w=3806920/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=14871MB, aggrb=253765KB/s, minb=253765KB/s, maxb=253765KB/s, mint=60007msec, maxt=60007msec

Disk stats (read/write):
  vda: ios=373/2418666, merge=0/14165, ticks=896/5007780, in_queue=5009012, util=83.23%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=28357: Thu Aug  8 17:48:40 2019
  read : io=4671.6MB, bw=79726KB/s, iops=19931, runt= 60002msec
    clat (usec): min=25, max=17184, avg=391.87, stdev=504.53
     lat (usec): min=25, max=17185, avg=393.42, stdev=505.54
    clat percentiles (usec):
     |  1.00th=[   53],  5.00th=[   77], 10.00th=[  100], 20.00th=[  145],
     | 30.00th=[  191], 40.00th=[  225], 50.00th=[  258], 60.00th=[  310],
     | 70.00th=[  386], 80.00th=[  524], 90.00th=[  796], 95.00th=[ 1080],
     | 99.00th=[ 2096], 99.50th=[ 3152], 99.90th=[ 6304], 99.95th=[ 8640],
     | 99.99th=[11584]
    bw (KB  /s): min= 3136, max=14952, per=12.49%, avg=9961.36, stdev=2158.45
    lat (usec) : 50=0.69%, 100=9.08%, 250=37.88%, 500=31.07%, 750=10.17%
    lat (usec) : 1000=5.03%
    lat (msec) : 2=5.00%, 4=0.77%, 10=0.29%, 20=0.02%
  cpu          : usr=2.09%, sys=5.15%, ctx=1204505, majf=0, minf=75
  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    : total=r=1195923/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=4671.6MB, aggrb=79725KB/s, minb=79725KB/s, maxb=79725KB/s, mint=60002msec, maxt=60002msec

Disk stats (read/write):
  vda: ios=1191412/11, merge=0/7, ticks=308900/0, in_queue=308564, util=88.02%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=28387: Thu Aug  8 17:50:41 2019
  write: io=721664KB, bw=12027KB/s, iops=3006, runt= 60003msec
    clat (usec): min=25, max=335859, avg=2651.82, stdev=12755.43
     lat (usec): min=26, max=335860, avg=2653.19, stdev=12756.13
    clat percentiles (usec):
     |  1.00th=[   40],  5.00th=[   49], 10.00th=[   58], 20.00th=[   75],
     | 30.00th=[   93], 40.00th=[  112], 50.00th=[  137], 60.00th=[  173],
     | 70.00th=[  229], 80.00th=[  334], 90.00th=[  740], 95.00th=[ 3824],
     | 99.00th=[69120], 99.50th=[80384], 99.90th=[113152], 99.95th=[144384],
     | 99.99th=[242688]
    bw (KB  /s): min=   39, max= 3136, per=12.60%, avg=1515.42, stdev=576.26
    lat (usec) : 50=5.03%, 100=28.72%, 250=38.87%, 500=13.86%, 750=3.64%
    lat (usec) : 1000=1.58%
    lat (msec) : 2=2.12%, 4=1.28%, 10=1.09%, 20=0.07%, 50=1.05%
    lat (msec) : 100=2.51%, 250=0.16%, 500=0.01%
  cpu          : usr=0.36%, sys=1.63%, ctx=365230, majf=0, minf=75
  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    : total=r=0/w=180416/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=721664KB, aggrb=12027KB/s, minb=12027KB/s, maxb=12027KB/s, mint=60003msec, maxt=60003msec

Disk stats (read/write):
  vda: ios=203/180493, merge=0/2174, ticks=116/52384, in_queue=52424, util=86.20%

Speed​​test

Retrieving speedtest.net configuration...
Testing from OVH SAS (5.196.205.67)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Orange (Paris) [1.88 km]: 8.941 ms
Testing download speed................................................................................
Download: 205.61 Mbit/s
Testing upload speed................................................................................................
Upload: 176.93 Mbit/s

AdminVPS からの仮想サーバー:

安価な仮想サーバーのテスト

  • ロシア(モスクワ)の設備。
  • VPS Lite 料金 (10 GB SSD / 1x 2.5 GHz CPU / 0.5 GB RAM / 100 Mbit/s、1 TB トラフィックで月額 90 ルーブル)。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 2.6.32-042stab136.1 -- #1 SMP Wed Feb 27 09:04:24 MSK 2019
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz (4999.6 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   15:53:14 up  3:18,  1 user,  load average: 0.59, 0.17, 0.05; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Fri Aug 30 2019 15:53:14 - 16:21:22
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       28170580.3 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3711.6 MWIPS (9.9 s, 7 samples)
Execl Throughput                               2982.4 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        225229.6 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           91046.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        888742.6 KBps  (30.0 s, 2 samples)
Pipe Throughput                              650654.6 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 138666.0 lps   (10.0 s, 7 samples)
Process Creation                               7152.4 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   5964.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    767.5 lpm   (60.0 s, 2 samples)
System Call Overhead                         511002.9 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   28170580.3   2413.9
Double-Precision Whetstone                       55.0       3711.6    674.8
Execl Throughput                                 43.0       2982.4    693.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     225229.6    568.8
File Copy 256 bufsize 500 maxblocks            1655.0      91046.0    550.1
File Copy 4096 bufsize 8000 maxblocks          5800.0     888742.6   1532.3
Pipe Throughput                               12440.0     650654.6    523.0
Pipe-based Context Switching                   4000.0     138666.0    346.7
Process Creation                                126.0       7152.4    567.7
Shell Scripts (1 concurrent)                     42.4       5964.2   1406.7
Shell Scripts (8 concurrent)                      6.0        767.5   1279.2
System Call Overhead                          15000.0     511002.9    340.7
                                                                   ========
System Benchmarks Index Score                                         754.8

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 2.94045 s, 730 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 3.55853 s, 603 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=14900: Fri Aug 30 16:22:34 2019
  read : io=54984MB, bw=938352KB/s, iops=234587, runt= 60003msec
    clat (usec): min=1, max=62031, avg=26.56, stdev=446.34
     lat (usec): min=1, max=62031, avg=26.96, stdev=450.31
    clat percentiles (usec):
     |  1.00th=[    1],  5.00th=[    2], 10.00th=[    2], 20.00th=[    2],
     | 30.00th=[    2], 40.00th=[    2], 50.00th=[    2], 60.00th=[    2],
     | 70.00th=[    2], 80.00th=[    3], 90.00th=[    3], 95.00th=[   31],
     | 99.00th=[  165], 99.50th=[  217], 99.90th=[ 7072], 99.95th=[ 9024],
     | 99.99th=[15040]
    bw (KB  /s): min=21456, max=174302, per=12.47%, avg=117027.11, stdev=42929.20
    lat (usec) : 2=4.60%, 4=89.15%, 10=1.10%, 20=0.01%, 50=1.05%
    lat (usec) : 100=0.35%, 250=3.34%, 500=0.16%, 750=0.02%, 1000=0.01%
    lat (msec) : 2=0.01%, 4=0.01%, 10=0.16%, 20=0.04%, 50=0.01%
    lat (msec) : 100=0.01%
  cpu          : usr=2.61%, sys=9.53%, ctx=620299, majf=0, minf=570
  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    : total=r=14075979/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=54984MB, aggrb=938351KB/s, minb=938351KB/s, maxb=938351KB/s, mint=60003msec, maxt=60003msec

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=14928: Fri Aug 30 16:24:40 2019
  write: io=36143MB, bw=616817KB/s, iops=154204, runt= 60002msec
    clat (usec): min=1, max=1303.1K, avg=21.52, stdev=949.49
     lat (usec): min=2, max=1303.2K, avg=21.87, stdev=954.55
    clat percentiles (usec):
     |  1.00th=[    2],  5.00th=[    3], 10.00th=[    3], 20.00th=[    3],
     | 30.00th=[    3], 40.00th=[    3], 50.00th=[    3], 60.00th=[    4],
     | 70.00th=[    4], 80.00th=[    4], 90.00th=[    4], 95.00th=[    5],
     | 99.00th=[    7], 99.50th=[   16], 99.90th=[ 4016], 99.95th=[12992],
     | 99.99th=[26240]
    bw (KB  /s): min=    0, max=115099, per=14.05%, avg=86648.06, stdev=26702.40
    lat (usec) : 2=0.01%, 4=51.15%, 10=48.21%, 20=0.22%, 50=0.15%
    lat (usec) : 100=0.05%, 250=0.04%, 500=0.01%, 750=0.01%, 1000=0.01%
    lat (msec) : 2=0.03%, 4=0.04%, 10=0.04%, 20=0.05%, 50=0.02%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
    lat (msec) : 2000=0.01%
  cpu          : usr=1.72%, sys=8.45%, ctx=144694, majf=0, minf=294
  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    : total=r=0/w=9252559/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=36143MB, aggrb=616816KB/s, minb=616816KB/s, maxb=616816KB/s, mint=60002msec, maxt=60002msec

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=14914: Fri Aug 30 16:23:39 2019
  read : io=59737MB, bw=995.59MB/s, iops=254865, runt= 60003msec
    clat (usec): min=1, max=72040, avg=24.00, stdev=599.88
     lat (usec): min=1, max=72040, avg=24.41, stdev=606.49
    clat percentiles (usec):
     |  1.00th=[    1],  5.00th=[    1], 10.00th=[    1], 20.00th=[    2],
     | 30.00th=[    2], 40.00th=[    2], 50.00th=[    2], 60.00th=[    2],
     | 70.00th=[    2], 80.00th=[    2], 90.00th=[    3], 95.00th=[    4],
     | 99.00th=[  151], 99.50th=[  173], 99.90th=[ 7008], 99.95th=[ 8032],
     | 99.99th=[37120]
    bw (KB  /s): min=19192, max=184384, per=12.46%, avg=127002.56, stdev=45462.78
    lat (usec) : 2=12.84%, 4=81.80%, 10=0.65%, 20=0.01%, 50=0.99%
    lat (usec) : 100=0.38%, 250=3.13%, 500=0.06%, 750=0.01%, 1000=0.01%
    lat (msec) : 2=0.01%, 4=0.01%, 10=0.10%, 20=0.01%, 50=0.02%
    lat (msec) : 100=0.01%
  cpu          : usr=2.67%, sys=9.58%, ctx=593780, majf=0, minf=593
  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    : total=r=15292680/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=59737MB, aggrb=995.59MB/s, minb=995.59MB/s, maxb=995.59MB/s, mint=60003msec, maxt=60003msec

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=14942: Fri Aug 30 16:25:41 2019
  write: io=39225MB, bw=669414KB/s, iops=167353, runt= 60003msec
    clat (usec): min=1, max=379827, avg=20.92, stdev=624.60
     lat (usec): min=2, max=379827, avg=21.34, stdev=634.38
    clat percentiles (usec):
     |  1.00th=[    3],  5.00th=[    3], 10.00th=[    3], 20.00th=[    3],
     | 30.00th=[    3], 40.00th=[    3], 50.00th=[    4], 60.00th=[    4],
     | 70.00th=[    4], 80.00th=[    4], 90.00th=[    5], 95.00th=[    6],
     | 99.00th=[    8], 99.50th=[   18], 99.90th=[ 5024], 99.95th=[12992],
     | 99.99th=[25984]
    bw (KB  /s): min=    0, max=116944, per=13.08%, avg=87547.30, stdev=21280.85
    lat (usec) : 2=0.01%, 4=45.51%, 10=53.73%, 20=0.29%, 50=0.17%
    lat (usec) : 100=0.07%, 250=0.04%, 500=0.02%, 750=0.01%, 1000=0.01%
    lat (msec) : 2=0.02%, 4=0.03%, 10=0.04%, 20=0.05%, 50=0.02%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%
  cpu          : usr=2.02%, sys=8.87%, ctx=143552, majf=0, minf=292
  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    : total=r=0/w=10041717/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=39225MB, aggrb=669414KB/s, minb=669414KB/s, maxb=669414KB/s, mint=60003msec, maxt=60003msec

Speed​​test

Retrieving speedtest.net configuration...
Testing from AdminVPS LLC (45.128.204.77)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Megafon (Moscow) [1.61 km]: 9.505 ms
Testing download speed................................................................................
Download: 46.59 Mbit/s
Testing upload speed................................................................................................
Upload: 223.60 Mbit/s

FirstVDS の仮想サーバー:

安価な仮想サーバーのテスト

  • 機器はヒムキにある独自の WebDC DC にあります。
  • Razmka 料金 (10 GB HDD / 1x 1.6 GHz CPU / 0.5 GB RAM / 100 Mbit/s で月額 90 ルーブル);
  • アカウントごとに所定の料金を適用できるサーバーは XNUMX 台までに制限されます。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 2.6.32-042stab138.1 -- #1 SMP Wed May 15 09:33:10 MSK 2019
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: Intel(R) Xeon(R) CPU E5645 @ 2.40GHz (4799.4 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   11:09:19 up 6 min,  1 user,  load average: 0.63, 0.26, 0.09; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Thu Sep 19 2019 11:09:19 - 11:38:17
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       25600667.1 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3365.6 MWIPS (9.7 s, 7 samples)
Execl Throughput                               2971.8 lps   (29.4 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        361593.6 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           95868.2 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        940425.2 KBps  (30.0 s, 2 samples)
Pipe Throughput                              688570.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 143289.6 lps   (10.0 s, 7 samples)
Process Creation                               7362.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   5899.8 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    792.8 lpm   (60.1 s, 2 samples)
System Call Overhead                         548701.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   25600667.1   2193.7
Double-Precision Whetstone                       55.0       3365.6    611.9
Execl Throughput                                 43.0       2971.8    691.1
File Copy 1024 bufsize 2000 maxblocks          3960.0     361593.6    913.1
File Copy 256 bufsize 500 maxblocks            1655.0      95868.2    579.3
File Copy 4096 bufsize 8000 maxblocks          5800.0     940425.2   1621.4
Pipe Throughput                               12440.0     688570.1    553.5
Pipe-based Context Switching                   4000.0     143289.6    358.2
Process Creation                                126.0       7362.8    584.3
Shell Scripts (1 concurrent)                     42.4       5899.8   1391.5
Shell Scripts (8 concurrent)                      6.0        792.8   1321.4
System Call Overhead                          15000.0     548701.6    365.8
                                                                   ========
System Benchmarks Index Score                                         793.2

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 2.6182 s, 820 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 3.26521 s, 658 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=12163: Thu Sep 19 11:39:42 2019
  read : io=114900KB, bw=1913.3KB/s, iops=478, runt= 60056msec
    clat (usec): min=457, max=260997, avg=16710.38, stdev=12073.24
     lat (usec): min=457, max=260998, avg=16710.77, stdev=12073.25
    clat percentiles (usec):
     |  1.00th=[ 1672],  5.00th=[ 3376], 10.00th=[ 5344], 20.00th=[ 9280],
     | 30.00th=[11968], 40.00th=[13888], 50.00th=[15168], 60.00th=[16064],
     | 70.00th=[18048], 80.00th=[20864], 90.00th=[27520], 95.00th=[36096],
     | 99.00th=[63232], 99.50th=[84480], 99.90th=[120320], 99.95th=[126464],
     | 99.99th=[203776]
    bw (KB  /s): min=   77, max=  974, per=12.53%, avg=239.67, stdev=72.01
    lat (usec) : 500=0.01%, 750=0.03%, 1000=0.09%
    lat (msec) : 2=1.55%, 4=4.97%, 10=15.51%, 20=55.45%, 50=20.53%
    lat (msec) : 100=1.50%, 250=0.37%, 500=0.01%
  cpu          : usr=0.07%, sys=0.21%, ctx=55467, majf=0, minf=337
  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    : total=r=28725/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=114900KB, aggrb=1913KB/s, minb=1913KB/s, maxb=1913KB/s, mint=60056msec, maxt=60056msec

Disk stats (read/write):
  ploop64982p1: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=12197: Thu Sep 19 11:41:57 2019
  write: io=1083.5MB, bw=18467KB/s, iops=4616, runt= 60076msec
    clat (usec): min=2, max=6926.2K, avg=367.68, stdev=21842.99
     lat (usec): min=2, max=6926.2K, avg=367.90, stdev=21843.36
    clat percentiles (usec):
     |  1.00th=[    3],  5.00th=[    3], 10.00th=[    3], 20.00th=[    3],
     | 30.00th=[    3], 40.00th=[    4], 50.00th=[    4], 60.00th=[    4],
     | 70.00th=[    5], 80.00th=[    5], 90.00th=[    6], 95.00th=[    8],
     | 99.00th=[ 4192], 99.50th=[ 7904], 99.90th=[69120], 99.95th=[123392],
     | 99.99th=[399360]
    bw (KB  /s): min=    0, max=80926, per=32.72%, avg=6041.82, stdev=14957.11
    lat (usec) : 4=34.19%, 10=61.97%, 20=2.06%, 50=0.22%, 100=0.08%
    lat (usec) : 250=0.03%, 500=0.01%, 750=0.02%, 1000=0.01%
    lat (msec) : 2=0.06%, 4=0.27%, 10=0.71%, 20=0.20%, 50=0.05%
    lat (msec) : 100=0.03%, 250=0.08%, 500=0.01%, 750=0.01%, 1000=0.01%
    lat (msec) : 2000=0.01%, >=2000=0.01%
  cpu          : usr=0.11%, sys=0.29%, ctx=9814, majf=0, minf=316
  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    : total=r=0/w=277362/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=1083.5MB, aggrb=18467KB/s, minb=18467KB/s, maxb=18467KB/s, mint=60076msec, maxt=60076msec

Disk stats (read/write):
  ploop64982p1: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=12180: Thu Sep 19 11:40:56 2019
  read : io=116792KB, bw=1946.3KB/s, iops=486, runt= 60010msec
    clat (usec): min=661, max=214319, avg=16434.78, stdev=9799.39
     lat (usec): min=662, max=214320, avg=16435.17, stdev=9799.40
    clat percentiles (msec):
     |  1.00th=[    3],  5.00th=[    5], 10.00th=[    7], 20.00th=[   11],
     | 30.00th=[   13], 40.00th=[   15], 50.00th=[   16], 60.00th=[   17],
     | 70.00th=[   18], 80.00th=[   21], 90.00th=[   27], 95.00th=[   35],
     | 99.00th=[   53], 99.50th=[   61], 99.90th=[   85], 99.95th=[  102],
     | 99.99th=[  157]
    bw (KB  /s): min=   63, max=  608, per=12.50%, avg=243.27, stdev=57.65
    lat (usec) : 750=0.01%, 1000=0.02%
    lat (msec) : 2=0.96%, 4=3.90%, 10=14.93%, 20=58.60%, 50=20.37%
    lat (msec) : 100=1.15%, 250=0.06%
  cpu          : usr=0.07%, sys=0.22%, ctx=59042, majf=0, minf=345
  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    : total=r=29198/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=116792KB, aggrb=1946KB/s, minb=1946KB/s, maxb=1946KB/s, mint=60010msec, maxt=60010msec

Disk stats (read/write):
  ploop64982p1: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=12214: Thu Sep 19 11:43:02 2019
  write: io=9637.8MB, bw=156250KB/s, iops=39062, runt= 63162msec
    clat (usec): min=2, max=3426.4K, avg=52.82, stdev=5360.24
     lat (usec): min=2, max=3431.9K, avg=53.16, stdev=5364.23
    clat percentiles (usec):
     |  1.00th=[    2],  5.00th=[    3], 10.00th=[    3], 20.00th=[    3],
     | 30.00th=[    3], 40.00th=[    3], 50.00th=[    3], 60.00th=[    3],
     | 70.00th=[    4], 80.00th=[    4], 90.00th=[    4], 95.00th=[    5],
     | 99.00th=[    9], 99.50th=[   26], 99.90th=[10048], 99.95th=[12224],
     | 99.99th=[50944]
    bw (KB  /s): min=    0, max=128798, per=27.35%, avg=42731.62, stdev=35978.01
    lat (usec) : 4=67.85%, 10=31.28%, 20=0.31%, 50=0.16%, 100=0.04%
    lat (usec) : 250=0.02%, 500=0.01%, 750=0.02%, 1000=0.01%
    lat (msec) : 2=0.03%, 4=0.04%, 10=0.14%, 20=0.07%, 50=0.02%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
    lat (msec) : 2000=0.01%, >=2000=0.01%
  cpu          : usr=0.73%, sys=2.03%, ctx=81357, majf=0, minf=314
  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    : total=r=0/w=2467265/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=9637.8MB, aggrb=156249KB/s, minb=156249KB/s, maxb=156249KB/s, mint=63162msec, maxt=63162msec

Disk stats (read/write):
  ploop64982p1: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%

Speed​​test

Retrieving speedtest.net configuration...
Testing from JSC IOT (62.109.10.90)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by RETN (Moscow) [0.12 km]: 2.969 ms
Testing download speed................................................................................
Download: 88.09 Mbit/s
Testing upload speed................................................................................................
Upload: 85.93 Mbit/s

VPSHoster からの仮想サーバー:

安価な仮想サーバーのテスト

  • 機器はロシア (モスクワ、DC マロスネット) とラトビア (残念ながらテストは行われていません) にあります。
  • Cloud Pico 料金表 (5 GB SSD / 1x 2.26 GHz CPU / 0.5 GB RAM / 10 Mbit/s で月額 99 ルーブル)。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-9-amd64 -- #1 SMP Debian 4.9.168-1+deb9u5 (2019-08-11)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version 2.5+ (4533.5 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   19:01:21 up 5 min,  1 user,  load average: 0.82, 0.31, 0.11; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Fri Aug 30 2019 19:01:21 - 19:29:52
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       21765832.2 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2833.6 MWIPS (9.6 s, 7 samples)
Execl Throughput                               2180.1 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        170044.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           44415.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        529105.8 KBps  (30.0 s, 2 samples)
Pipe Throughput                              245395.9 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  54778.9 lps   (10.0 s, 7 samples)
Process Creation                               5248.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4598.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    607.0 lpm   (60.0 s, 2 samples)
System Call Overhead                         214860.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   21765832.2   1865.1
Double-Precision Whetstone                       55.0       2833.6    515.2
Execl Throughput                                 43.0       2180.1    507.0
File Copy 1024 bufsize 2000 maxblocks          3960.0     170044.4    429.4
File Copy 256 bufsize 500 maxblocks            1655.0      44415.3    268.4
File Copy 4096 bufsize 8000 maxblocks          5800.0     529105.8    912.3
Pipe Throughput                               12440.0     245395.9    197.3
Pipe-based Context Switching                   4000.0      54778.9    136.9
Process Creation                                126.0       5248.8    416.6
Shell Scripts (1 concurrent)                     42.4       4598.2   1084.5
Shell Scripts (8 concurrent)                      6.0        607.0   1011.7
System Call Overhead                          15000.0     214860.1    143.2
                                                                   ========
System Benchmarks Index Score                                         460.3

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 100.5 s, 21.4 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 98.6091 s, 21.8 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=1656: Fri Aug 30 19:35:36 2019
  read : io=3989.2MB, bw=68076KB/s, iops=17019, runt= 60004msec
    clat (usec): min=5, max=136069, avg=455.71, stdev=570.96
     lat (usec): min=7, max=136071, avg=457.93, stdev=573.08
    clat percentiles (usec):
     |  1.00th=[  183],  5.00th=[  249], 10.00th=[  286], 20.00th=[  330],
     | 30.00th=[  362], 40.00th=[  394], 50.00th=[  418], 60.00th=[  446],
     | 70.00th=[  482], 80.00th=[  524], 90.00th=[  620], 95.00th=[  724],
     | 99.00th=[ 1240], 99.50th=[ 1608], 99.90th=[ 2896], 99.95th=[ 5088],
     | 99.99th=[15808]
    bw (KB  /s): min= 4664, max= 9752, per=12.50%, avg=8512.83, stdev=722.84
    lat (usec) : 10=0.03%, 20=0.20%, 50=0.27%, 100=0.05%, 250=4.56%
    lat (usec) : 500=69.72%, 750=20.87%, 1000=2.80%
    lat (msec) : 2=1.29%, 4=0.16%, 10=0.04%, 20=0.02%, 50=0.01%
    lat (msec) : 100=0.01%, 250=0.01%
  cpu          : usr=2.76%, sys=9.35%, ctx=1025023, majf=0, minf=74
  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    : total=r=1021210/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=3989.2MB, aggrb=68076KB/s, minb=68076KB/s, maxb=68076KB/s, mint=60004msec, maxt=60004msec

Disk stats (read/write):
  vda: ios=1013181/9, merge=0/4, ticks=110508/0, in_queue=109852, util=96.31%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=1694: Fri Aug 30 19:38:53 2019
  write: io=6403.8MB, bw=109257KB/s, iops=27314, runt= 60018msec
    clat (usec): min=6, max=924982, avg=215.01, stdev=3585.50
     lat (usec): min=8, max=924984, avg=227.42, stdev=3631.69
    clat percentiles (usec):
     |  1.00th=[    7],  5.00th=[    8], 10.00th=[    8], 20.00th=[    8],
     | 30.00th=[    8], 40.00th=[    9], 50.00th=[   12], 60.00th=[   16],
     | 70.00th=[   16], 80.00th=[   18], 90.00th=[   24], 95.00th=[   34],
     | 99.00th=[ 5984], 99.50th=[13888], 99.90th=[25984], 99.95th=[32640],
     | 99.99th=[103936]
    bw (KB  /s): min=    0, max=22274, per=13.14%, avg=14350.91, stdev=4988.36
    lat (usec) : 10=43.34%, 20=40.96%, 50=12.67%, 100=0.83%, 250=0.39%
    lat (usec) : 500=0.17%, 750=0.05%, 1000=0.04%
    lat (msec) : 2=0.13%, 4=0.22%, 10=0.48%, 20=0.48%, 50=0.21%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
  cpu          : usr=2.97%, sys=6.15%, ctx=93218, majf=0, minf=76
  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    : total=r=0/w=1639344/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=6403.8MB, aggrb=109256KB/s, minb=109256KB/s, maxb=109256KB/s, mint=60018msec, maxt=60018msec

Disk stats (read/write):
  vda: ios=162/1011414, merge=0/10885, ticks=4/2873892, in_queue=2878996, util=98.03%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=1681: Fri Aug 30 19:37:53 2019
  read : io=5401.3MB, bw=92171KB/s, iops=23042, runt= 60007msec
    clat (usec): min=59, max=78186, avg=331.60, stdev=200.65
     lat (usec): min=61, max=78187, avg=333.99, stdev=204.48
    clat percentiles (usec):
     |  1.00th=[  141],  5.00th=[  181], 10.00th=[  205], 20.00th=[  239],
     | 30.00th=[  266], 40.00th=[  290], 50.00th=[  318], 60.00th=[  342],
     | 70.00th=[  366], 80.00th=[  398], 90.00th=[  458], 95.00th=[  532],
     | 99.00th=[  756], 99.50th=[  900], 99.90th=[ 1496], 99.95th=[ 1848],
     | 99.99th=[ 6752]
    bw (KB  /s): min= 7624, max=12520, per=12.50%, avg=11521.79, stdev=787.53
    lat (usec) : 100=0.06%, 250=23.92%, 500=69.50%, 750=5.51%, 1000=0.67%
    lat (msec) : 2=0.31%, 4=0.02%, 10=0.01%, 20=0.01%, 50=0.01%
    lat (msec) : 100=0.01%
  cpu          : usr=3.75%, sys=8.62%, ctx=1388538, majf=0, minf=66
  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    : total=r=1382727/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=5401.3MB, aggrb=92171KB/s, minb=92171KB/s, maxb=92171KB/s, mint=60007msec, maxt=60007msec

Disk stats (read/write):
  vda: ios=1381633/8, merge=0/4, ticks=134112/0, in_queue=133288, util=98.91%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=1707: Fri Aug 30 19:40:01 2019
  write: io=794504KB, bw=13239KB/s, iops=3309, runt= 60012msec
    clat (usec): min=68, max=49039, avg=2399.92, stdev=7968.19
     lat (usec): min=70, max=49041, avg=2402.85, stdev=7968.29
    clat percentiles (usec):
     |  1.00th=[  102],  5.00th=[  131], 10.00th=[  153], 20.00th=[  183],
     | 30.00th=[  217], 40.00th=[  255], 50.00th=[  290], 60.00th=[  326],
     | 70.00th=[  362], 80.00th=[  414], 90.00th=[  524], 95.00th=[28800],
     | 99.00th=[36608], 99.50th=[38656], 99.90th=[41216], 99.95th=[42240],
     | 99.99th=[44288]
    bw (KB  /s): min= 1050, max= 3260, per=12.50%, avg=1655.20, stdev=381.52
    lat (usec) : 100=0.85%, 250=37.75%, 500=50.33%, 750=4.27%, 1000=0.07%
    lat (msec) : 2=0.02%, 4=0.01%, 10=0.01%, 20=0.01%, 50=6.69%
  cpu          : usr=0.71%, sys=3.17%, ctx=411519, majf=0, minf=58
  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    : total=r=0/w=198626/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=794504KB, aggrb=13239KB/s, minb=13239KB/s, maxb=13239KB/s, mint=60012msec, maxt=60012msec

Disk stats (read/write):
  vda: ios=163/198371, merge=0/2029, ticks=36/67048, in_queue=66932, util=82.45%

Speed​​test

Retrieving speedtest.net configuration...
Testing from Tender Consult LLC (45.15.253.158)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by ISP Balnet (Balashikha) [2.50 km]: 13.869 ms
Testing download speed................................................................................
Download: 72.68 Mbit/s
Testing upload speed................................................................................................
Upload: 83.40 Mbit/s

MyVPS からの仮想サーバー:

安価な仮想サーバーのテスト

  • 機器はロシア (モスクワ) にあります。
  • 料金表 OVZ-256 (10 GB HDD+SSD / 1x 0.6 GHz CPU / 0.25 GB RAM / 10 Mbit/s、月額 99 ルーブル)。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 3.16.0-042stab125.5 -- #1 SMP Tue Oct 17 12:48:22 MSK 2017
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz (5199.8 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   16:31:09 up 4 min,  1 user,  load average: 1.15, 0.58, 0.22; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Fri Aug 30 2019 16:31:09 - 17:16:10
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables        7193473.9 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3936.8 MWIPS (10.0 s, 7 samples)
Execl Throughput                                925.1 lps   (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        132131.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           31146.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        311906.1 KBps  (30.0 s, 2 samples)
Pipe Throughput                              302839.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  56423.7 lps   (10.0 s, 7 samples)
Process Creation                               2331.4 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   1713.8 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    231.8 lpm   (60.2 s, 2 samples)
System Call Overhead                         606956.9 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    7193473.9    616.4
Double-Precision Whetstone                       55.0       3936.8    715.8
Execl Throughput                                 43.0        925.1    215.1
File Copy 1024 bufsize 2000 maxblocks          3960.0     132131.0    333.7
File Copy 256 bufsize 500 maxblocks            1655.0      31146.0    188.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     311906.1    537.8
Pipe Throughput                               12440.0     302839.1    243.4
Pipe-based Context Switching                   4000.0      56423.7    141.1
Process Creation                                126.0       2331.4    185.0
Shell Scripts (1 concurrent)                     42.4       1713.8    404.2
Shell Scripts (8 concurrent)                      6.0        231.8    386.3
System Call Overhead                          15000.0     606956.9    404.6
                                                                   ========
System Benchmarks Index Score                                         322.9

dd1Mのブロックで

2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 17.4143 s, 123 MB/s

64Kブロック

32768+0 records in
32768+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 28.5987 s, 75.1 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=26682: Fri Aug 30 17:18:34 2019
  read : io=3040.5MB, bw=51874KB/s, iops=12968, runt= 60018msec
    clat (usec): min=1, max=1892.9K, avg=614.31, stdev=6031.40
     lat (usec): min=1, max=1892.9K, avg=614.48, stdev=6031.40
    clat percentiles (usec):
     |  1.00th=[    1],  5.00th=[  125], 10.00th=[  187], 20.00th=[  237],
     | 30.00th=[  266], 40.00th=[  298], 50.00th=[  326], 60.00th=[  362],
     | 70.00th=[  402], 80.00th=[  454], 90.00th=[  556], 95.00th=[  692],
     | 99.00th=[ 6560], 99.50th=[17024], 99.90th=[40704], 99.95th=[48896],
     | 99.99th=[123392]
    bw (KB  /s): min=    4, max=18376, per=12.81%, avg=6645.61, stdev=3143.86
    lat (usec) : 2=1.27%, 4=2.58%, 10=0.02%, 20=0.01%, 50=0.01%
    lat (usec) : 100=0.31%, 250=20.10%, 500=61.29%, 750=10.17%, 1000=1.43%
    lat (msec) : 2=1.21%, 4=0.36%, 10=0.50%, 20=0.30%, 50=0.39%
    lat (msec) : 100=0.03%, 250=0.01%, 500=0.01%, 750=0.01%, 2000=0.01%
  cpu          : usr=0.53%, sys=2.04%, ctx=771415, majf=2, minf=391
  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    : total=r=778349/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=3040.5MB, aggrb=51874KB/s, minb=51874KB/s, maxb=51874KB/s, mint=60018msec, maxt=60018msec

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=26747: Fri Aug 30 17:21:33 2019
  write: io=244736KB, bw=4053.5KB/s, iops=1013, runt= 60377msec
    clat (usec): min=1, max=18855K, avg=1726.88, stdev=105186.89
     lat (usec): min=1, max=18855K, avg=1727.03, stdev=105186.90
    clat percentiles (usec):
     |  1.00th=[    2],  5.00th=[    3], 10.00th=[    3], 20.00th=[    4],
     | 30.00th=[    4], 40.00th=[    4], 50.00th=[    4], 60.00th=[    5],
     | 70.00th=[    6], 80.00th=[   10], 90.00th=[  724], 95.00th=[  868],
     | 99.00th=[50944], 99.50th=[51456], 99.90th=[54016], 99.95th=[83456],
     | 99.99th=[370688]
    bw (KB  /s): min=    0, max=36376, per=47.16%, avg=1911.33, stdev=4411.56
    lat (usec) : 2=0.04%, 4=18.28%, 10=61.11%, 20=3.07%, 50=0.47%
    lat (usec) : 100=0.12%, 250=0.01%, 500=1.47%, 750=6.52%, 1000=6.02%
    lat (msec) : 2=0.67%, 4=0.17%, 10=0.25%, 20=0.15%, 50=0.28%
    lat (msec) : 100=1.36%, 250=0.01%, 500=0.01%, 1000=0.01%, 2000=0.01%
    lat (msec) : >=2000=0.01%
  cpu          : usr=0.02%, sys=0.19%, ctx=100164, majf=0, minf=303
  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    : total=r=0/w=61184/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=244736KB, aggrb=4053KB/s, minb=4053KB/s, maxb=4053KB/s, mint=60377msec, maxt=60377msec

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=26716: Fri Aug 30 17:20:18 2019
  read : io=3701.1MB, bw=62630KB/s, iops=15657, runt= 60527msec
    clat (usec): min=70, max=3091.9K, avg=504.94, stdev=11013.15
     lat (usec): min=70, max=3091.9K, avg=505.10, stdev=11013.15
    clat percentiles (usec):
     |  1.00th=[  135],  5.00th=[  189], 10.00th=[  221], 20.00th=[  258],
     | 30.00th=[  290], 40.00th=[  314], 50.00th=[  342], 60.00th=[  370],
     | 70.00th=[  398], 80.00th=[  442], 90.00th=[  516], 95.00th=[  612],
     | 99.00th=[ 1336], 99.50th=[ 3664], 99.90th=[21120], 99.95th=[29312],
     | 99.99th=[94720]
    bw (KB  /s): min=    2, max=17424, per=13.84%, avg=8669.37, stdev=3103.57
    lat (usec) : 100=0.18%, 250=17.35%, 500=70.81%, 750=9.09%, 1000=1.12%
    lat (msec) : 2=0.73%, 4=0.24%, 10=0.22%, 20=0.15%, 50=0.09%
    lat (msec) : 100=0.01%, 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
    lat (msec) : 2000=0.01%, >=2000=0.01%
  cpu          : usr=0.63%, sys=1.75%, ctx=954817, majf=0, minf=325
  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    : total=r=947695/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=3701.1MB, aggrb=62629KB/s, minb=62629KB/s, maxb=62629KB/s, mint=60527msec, maxt=60527msec

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=32
...
fio-2.8
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=26770: Fri Aug 30 17:22:35 2019
  write: io=424852KB, bw=7080.2KB/s, iops=1770, runt= 60006msec
    clat (usec): min=228, max=2046.4K, avg=859.72, stdev=18891.70
     lat (usec): min=228, max=2046.5K, avg=859.99, stdev=18891.83
    clat percentiles (usec):
     |  1.00th=[  278],  5.00th=[  294], 10.00th=[  310], 20.00th=[  326],
     | 30.00th=[  342], 40.00th=[  358], 50.00th=[  378], 60.00th=[  410],
     | 70.00th=[  446], 80.00th=[  482], 90.00th=[  532], 95.00th=[  604],
     | 99.00th=[ 1736], 99.50th=[ 5664], 99.90th=[37632], 99.95th=[114176],
     | 99.99th=[1056768]
    bw (KB  /s): min=    3, max= 5368, per=14.79%, avg=1047.15, stdev=707.25
    lat (usec) : 250=0.02%, 500=84.01%, 750=12.93%, 1000=0.95%
    lat (msec) : 2=1.23%, 4=0.26%, 10=0.26%, 20=0.16%, 50=0.09%
    lat (msec) : 100=0.02%, 250=0.02%, 500=0.01%, 750=0.02%, 1000=0.01%
    lat (msec) : 2000=0.01%, >=2000=0.01%
  cpu          : usr=0.08%, sys=0.43%, ctx=213556, majf=0, minf=308
  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    : total=r=0/w=106213/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=424852KB, aggrb=7080KB/s, minb=7080KB/s, maxb=7080KB/s, mint=60006msec, maxt=60006msec

Speed​​test

Retrieving speedtest.net configuration...
Testing from OOO National Telecommunications (109.120.149.95)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Rostelecom (Moscow) [1.61 km]: 11.253 ms
Testing download speed................................................................................
Download: 687.09 Mbit/s
Testing upload speed................................................................................................
Upload: 10.49 Mbit/s

IHOR の仮想サーバー:

安価な仮想サーバーのテスト

  • 機器はロシア (モスクワ、DC マロスネット) にあります。
  • Minimus (HDD) 料金表 (10 GB HDD / 1x 2.4 GHz CPU / 0.5 GB RAM / 100 Mbit/s で月額 100 ルーブル)。

テスト:

Unixベンチ

   BYTE UNIX Benchmarks (Version 5.1.3)

   System: test: GNU/Linux
   OS: GNU/Linux -- 4.9.0-4-amd64 -- #1 SMP Debian 4.9.51-1 (2017-09-28)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (4800.2 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   16:23:39 up 390 days, 18:17,  1 user,  load average: 1.28, 1.21, 0.67; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Пн июл 15 2019 16:23:39 - 16:52:26
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       22433265.9 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2999.0 MWIPS (10.0 s, 7 samples)
Execl Throughput                               3246.7 lps   (29.3 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        681169.7 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          201657.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1275322.6 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1639916.4 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 210397.4 lps   (10.0 s, 7 samples)
Process Creation                               7607.2 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   6396.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    810.7 lpm   (60.0 s, 2 samples)
System Call Overhead                        2527110.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   22433265.9   1922.3
Double-Precision Whetstone                       55.0       2999.0    545.3
Execl Throughput                                 43.0       3246.7    755.0
File Copy 1024 bufsize 2000 maxblocks          3960.0     681169.7   1720.1
File Copy 256 bufsize 500 maxblocks            1655.0     201657.4   1218.5
File Copy 4096 bufsize 8000 maxblocks          5800.0    1275322.6   2198.8
Pipe Throughput                               12440.0    1639916.4   1318.3
Pipe-based Context Switching                   4000.0     210397.4    526.0
Process Creation                                126.0       7607.2    603.7
Shell Scripts (1 concurrent)                     42.4       6396.9   1508.7
Shell Scripts (8 concurrent)                      6.0        810.7   1351.1
System Call Overhead                          15000.0    2527110.6   1684.7
                                                                   ========
System Benchmarks Index Score                                        1148.1

dd1Mのブロックで

2048+0 записей получено
2048+0 записей отправлено
2147483648 байт (2,1 GB, 2,0 GiB) скопирован, 33,811 s, 63,5 MB/s

64Kブロック

32768+0 записей получено
32768+0 записей отправлено
2147483648 байт (2,1 GB, 2,0 GiB) скопирован, 37,3774 s, 57,5 MB/s

FIO読む

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=32063: Mon Jul 15 16:55:28 2019
  read : io=108176KB, bw=1802.6KB/s, iops=450, runt= 60013msec
    clat (usec): min=82, max=28243, avg=17743.49, stdev=966.00
     lat (usec): min=83, max=28244, avg=17744.08, stdev=965.98
    clat percentiles (usec):
     |  1.00th=[16512],  5.00th=[17280], 10.00th=[17280], 20.00th=[17536],
     | 30.00th=[17536], 40.00th=[17536], 50.00th=[17536], 60.00th=[17792],
     | 70.00th=[17792], 80.00th=[18304], 90.00th=[18560], 95.00th=[18560],
     | 99.00th=[18816], 99.50th=[20096], 99.90th=[23168], 99.95th=[24192],
     | 99.99th=[26752]
    lat (usec) : 100=0.01%, 250=0.10%, 500=0.05%, 750=0.01%
    lat (msec) : 4=0.01%, 10=0.03%, 20=99.26%, 50=0.53%
  cpu          : usr=0.07%, sys=0.29%, ctx=27079, majf=0, minf=73
  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    : total=r=27044/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=108176KB, aggrb=1802KB/s, minb=1802KB/s, maxb=1802KB/s, mint=60013msec, maxt=60013msec

Disk stats (read/write):
  vda: ios=26973/11, merge=0/13, ticks=477280/52, in_queue=477412, util=99.99%

書きます

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes
benchmark: Laying out IO file(s) (1 file(s) / 256MB)

benchmark: (groupid=0, jobs=8): err= 0: pid=32085: Mon Jul 15 16:57:39 2019
  write: io=143120KB, bw=2380.6KB/s, iops=595, runt= 60121msec
    clat (usec): min=2, max=6778.8K, avg=11090.86, stdev=180942.60
     lat (usec): min=2, max=6778.8K, avg=11091.21, stdev=180943.13
    clat percentiles (usec):
     |  1.00th=[    3],  5.00th=[    5], 10.00th=[    8], 20.00th=[    8],
     | 30.00th=[    9], 40.00th=[   10], 50.00th=[   11], 60.00th=[   12],
     | 70.00th=[   14], 80.00th=[   18], 90.00th=[   25], 95.00th=[   44],
     | 99.00th=[115200], 99.50th=[183296], 99.90th=[3489792], 99.95th=[4620288],
     | 99.99th=[6062080]
    lat (usec) : 4=2.42%, 10=37.29%, 20=44.42%, 50=11.45%, 100=0.93%
    lat (usec) : 250=0.08%, 500=0.02%, 750=0.01%, 1000=0.01%
    lat (msec) : 2=0.03%, 4=0.03%, 10=0.15%, 20=0.44%, 50=1.22%
    lat (msec) : 100=0.33%, 250=0.80%, 500=0.02%, 750=0.01%, 1000=0.04%
    lat (msec) : 2000=0.15%, >=2000=0.18%
  cpu          : usr=0.02%, sys=0.27%, ctx=22948, majf=0, minf=56
  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    : total=r=0/w=35780/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=143120KB, aggrb=2380KB/s, minb=2380KB/s, maxb=2380KB/s, mint=60121msec, maxt=60121msec

Disk stats (read/write):
  vda: ios=0/18270, merge=0/9732, ticks=0/124272, in_queue=124652, util=98.72%

直読

benchmark: (g=0): rw=randread, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes

benchmark: (groupid=0, jobs=8): err= 0: pid=32074: Mon Jul 15 16:56:29 2019
  read : io=108200KB, bw=1803.2KB/s, iops=450, runt= 60011msec
    clat (usec): min=55, max=27740, avg=17737.39, stdev=906.38
     lat (usec): min=56, max=27741, avg=17738.01, stdev=906.37
    clat percentiles (usec):
     |  1.00th=[16768],  5.00th=[17280], 10.00th=[17280], 20.00th=[17536],
     | 30.00th=[17536], 40.00th=[17536], 50.00th=[17536], 60.00th=[17792],
     | 70.00th=[17792], 80.00th=[18304], 90.00th=[18560], 95.00th=[18560],
     | 99.00th=[18816], 99.50th=[19072], 99.90th=[21120], 99.95th=[22912],
     | 99.99th=[25728]
    lat (usec) : 100=0.06%, 250=0.06%, 500=0.01%, 750=0.02%, 1000=0.01%
    lat (msec) : 2=0.01%, 4=0.03%, 10=0.03%, 20=99.59%, 50=0.20%
  cpu          : usr=0.08%, sys=0.21%, ctx=27079, majf=0, minf=65
  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    : total=r=27050/w=0/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
   READ: io=108200KB, aggrb=1803KB/s, minb=1803KB/s, maxb=1803KB/s, mint=60011msec, maxt=60011msec

Disk stats (read/write):
  vda: ios=27003/11, merge=0/11, ticks=477000/360, in_queue=477400, util=99.97%

直接書き込み

benchmark: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=psync, iodepth=32
...
fio-2.16
Starting 8 processes

benchmark: (groupid=0, jobs=8): err= 0: pid=32097: Mon Jul 15 16:58:40 2019
  write: io=49228KB, bw=839821B/s, iops=205, runt= 60024msec
    clat (usec): min=104, max=543532, avg=26720.27, stdev=101994.62
     lat (usec): min=105, max=543533, avg=26721.00, stdev=101994.62
    clat percentiles (msec):
     |  1.00th=[    3],  5.00th=[    3], 10.00th=[    4], 20.00th=[    4],
     | 30.00th=[    4], 40.00th=[    4], 50.00th=[    4], 60.00th=[    4],
     | 70.00th=[    4], 80.00th=[    5], 90.00th=[    5], 95.00th=[  235],
     | 99.00th=[  498], 99.50th=[  515], 99.90th=[  529], 99.95th=[  537],
     | 99.99th=[  545]
    lat (usec) : 250=0.17%, 500=0.02%, 1000=0.03%
    lat (msec) : 2=0.20%, 4=72.51%, 10=21.99%, 20=0.02%, 50=0.01%
    lat (msec) : 100=0.02%, 250=0.04%, 500=4.04%, 750=0.96%
  cpu          : usr=0.03%, sys=0.17%, ctx=29841, majf=0, minf=75
  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    : total=r=0/w=12307/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: io=49228KB, aggrb=820KB/s, minb=820KB/s, maxb=820KB/s, mint=60024msec, maxt=60024msec

Disk stats (read/write):
  vda: ios=0/18000, merge=0/180, ticks=0/204144, in_queue=203652, util=99.65%

Speed​​test

Retrieving speedtest.net configuration...
Testing from NTX Technologies s.r.o. (194.67.207.200)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Megafon (Moscow) [1.61 km]: 9.079 ms
Testing download speed................................................................................
Download: 102.30 Mbit/s
Testing upload speed................................................................................................
Upload: 113.43 Mbit/s

結局何が起こったのか(写真はクリック可能です):

オウム Unixベンチ

安価な仮想サーバーのテスト

読み取り/書き込み速度 dd (MB/s)

1M

安価な仮想サーバーのテスト

64K

安価な仮想サーバーのテスト

さまざまな動作モードでの IOPS の数 fio

ランダム読み取り

安価な仮想サーバーのテスト

ランダム書き込み

安価な仮想サーバーのテスト

ランダムな直接読み取り

安価な仮想サーバーのテスト

ランダム直接書き込み

安価な仮想サーバーのテスト

スピードテストによるネットワーク速度 (Mbit/s)

安価な仮想サーバーのテスト
青 - アップロード、赤 - ダウンロード。 最後の XNUMX つのホスティング会社のアップロード速度が遅いのは、チャネルの制限によって説明されます。

結論は出しませんが、高価で遅いサーバーよりも、安価で生産性の高いサーバーの方が優れているとだけ言っておきます。

エフゲニーに感謝します(やあ)Lite.Host から、テスト用のスクリプトの準備に役立ちます。

登録ユーザーのみがアンケートに参加できます。 ログインお願いします。

最近、多くのホスティング会社が、アカウントごとに XNUMX 台のサーバーという同じ種類の制限を設けた非常に安価な料金プランを作成し始めています。 このような制限のある VPS/VDS に興味がありますか?

  • 視聴者の38%がはい、複数必要ありません159

  • 視聴者の38%がはい、いくつかのアカウントを作成します58

  • 視聴者の38%がいいえ、そのような制限では購入しません51

268 人のユーザーが投票しました。 62名のユーザーが棄権した。

出所: habr.com

コメントを追加します