シン・VPSを使ってみた

「次世代のVPS」の名を関したシン・VPS。その名の通り一般向け国内事業者としては珍しい10Gbps共有をカタログスペックとして挙げている。

実際のところはどうなのか。

VPS 4GBで検証。最後に書いたが実質200円/月ぐらいで使えるCPをやってたので先12ヶ月はとりあえず使う。

まずは手始めにスピードテスト

code:speedtest

# speedtest --server-id=24333

Speedtest by Ookla

Server: Rakuten Mobile, Inc - Tokyo (id: 24333)

ISP: XServer

Idle Latency: 11.35 ms (jitter: 1.76ms, low: 10.29ms, high: 14.06ms)

Download: 480.28 Mbps (data used: 739.7 MB)

1080.71 ms (jitter: 89.35ms, low: 10.48ms, high: 1567.00ms)

Upload: 98.50 Mbps (data used: 154.5 MB)

10.46 ms (jitter: 0.58ms, low: 10.03ms, high: 18.42ms)

Packet Loss: 0.0%

期待よりは全然出ない！

が下り500Mbpsと100Mbpsが当然のような顔をしている国内事業者にしては出る方。

上りはもうちょっと頑張ってほしい。同じ収容ホストにdevrandom無限送信してるやつとか居たりしないですか。

追記：全然頭打ちが変わらないため、収容ホストとかそういうのではなく明文化されていない制限に見える

上りは100Mbps制限がかかっていると考えるのが妥当な挙動

unixbenchも打ってみた

code:unixbench

========================================================================

BYTE UNIX Benchmarks (Version 5.1.3)

System: localhost: GNU/Linux

Machine: x86_64 (x86_64)

Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")

CPU 0: AMD EPYC-Milan Processor (3992.5 bogomips)

x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, AMD virtualization, SYSCALL/SYSRET

CPU 1: AMD EPYC-Milan Processor (3992.5 bogomips)

x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, AMD virtualization, SYSCALL/SYSRET

CPU 2: AMD EPYC-Milan Processor (3992.5 bogomips)

x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, AMD virtualization, SYSCALL/SYSRET

CPU 3: AMD EPYC-Milan Processor (3992.5 bogomips)

x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, AMD virtualization, SYSCALL/SYSRET

22:52:41 up 3:36, 3 users, load average: 0.00, 0.00, 0.00; runlevel 2024-03-20

------------------------------------------------------------------------

Benchmark Run: 水 3月 20 2024 22:52:41 - 23:20:41

4 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 42310417.3 lps (10.0 s, 7 samples)

Double-Precision Whetstone 7448.3 MWIPS (9.9 s, 7 samples)

Execl Throughput 4175.5 lps (30.0 s, 2 samples)

File Copy 1024 bufsize 2000 maxblocks 1227495.2 KBps (30.0 s, 2 samples)

File Copy 256 bufsize 500 maxblocks 321413.0 KBps (30.0 s, 2 samples)

File Copy 4096 bufsize 8000 maxblocks 3647734.9 KBps (30.0 s, 2 samples)

Pipe Throughput 1699603.1 lps (10.0 s, 7 samples)

Pipe-based Context Switching 152431.8 lps (10.0 s, 7 samples)

Process Creation 8691.2 lps (30.0 s, 2 samples)

Shell Scripts (1 concurrent) 13328.9 lpm (60.0 s, 2 samples)

Shell Scripts (8 concurrent) 3823.9 lpm (60.0 s, 2 samples)

System Call Overhead 1436138.0 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX

Dhrystone 2 using register variables 116700.0 42310417.3 3625.6

Double-Precision Whetstone 55.0 7448.3 1354.2

Execl Throughput 43.0 4175.5 971.1

File Copy 1024 bufsize 2000 maxblocks 3960.0 1227495.2 3099.7

File Copy 256 bufsize 500 maxblocks 1655.0 321413.0 1942.1

File Copy 4096 bufsize 8000 maxblocks 5800.0 3647734.9 6289.2

Pipe Throughput 12440.0 1699603.1 1366.2

Pipe-based Context Switching 4000.0 152431.8 381.1

Process Creation 126.0 8691.2 689.8

Shell Scripts (1 concurrent) 42.4 13328.9 3143.6

Shell Scripts (8 concurrent) 6.0 3823.9 6373.2

System Call Overhead 15000.0 1436138.0 957.4

========

System Benchmarks Index Score 1810.9

------------------------------------------------------------------------

Benchmark Run: 水 3月 20 2024 23:20:41 - 23:48:41

4 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables 167527111.8 lps (10.0 s, 7 samples)

Double-Precision Whetstone 29774.0 MWIPS (9.9 s, 7 samples)

Execl Throughput 8415.7 lps (30.0 s, 2 samples)

File Copy 1024 bufsize 2000 maxblocks 4807443.5 KBps (30.0 s, 2 samples)

File Copy 256 bufsize 500 maxblocks 1276249.9 KBps (30.0 s, 2 samples)

File Copy 4096 bufsize 8000 maxblocks 14374204.0 KBps (30.0 s, 2 samples)

Pipe Throughput 6808825.2 lps (10.0 s, 7 samples)

Pipe-based Context Switching 423283.1 lps (10.0 s, 7 samples)

Process Creation 29421.8 lps (30.0 s, 2 samples)

Shell Scripts (1 concurrent) 28223.9 lpm (60.0 s, 2 samples)

Shell Scripts (8 concurrent) 3981.8 lpm (60.0 s, 2 samples)

System Call Overhead 5749321.2 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX

Dhrystone 2 using register variables 116700.0 167527111.8 14355.4

Double-Precision Whetstone 55.0 29774.0 5413.4

Execl Throughput 43.0 8415.7 1957.1

File Copy 1024 bufsize 2000 maxblocks 3960.0 4807443.5 12140.0

File Copy 256 bufsize 500 maxblocks 1655.0 1276249.9 7711.5

File Copy 4096 bufsize 8000 maxblocks 5800.0 14374204.0 24783.1

Pipe Throughput 12440.0 6808825.2 5473.3

Pipe-based Context Switching 4000.0 423283.1 1058.2

Process Creation 126.0 29421.8 2335.1

Shell Scripts (1 concurrent) 42.4 28223.9 6656.6

Shell Scripts (8 concurrent) 6.0 3981.8 6636.4

System Call Overhead 15000.0 5749321.2 3832.9

========

System Benchmarks Index Score 5524.4

iperf3

TCP下り/TCP上り/UDP下り/UDP上り

code:iperf3

-----------------------------------------------------------

Server listening on 5201

-----------------------------------------------------------

- - - - - - - - - - - - - - - - - - - - - - - - -

-----------------------------------------------------------

Server listening on 5201

-----------------------------------------------------------

- - - - - - - - - - - - - - - - - - - - - - - - -

-----------------------------------------------------------

Server listening on 5201

-----------------------------------------------------------

- - - - - - - - - - - - - - - - - - - - - - - - -

-----------------------------------------------------------

Server listening on 5201

-----------------------------------------------------------

- - - - - - - - - - - - - - - - - - - - - - - - -

-----------------------------------------------------------

Server listening on 5201

-----------------------------------------------------------

価格帯は他よりは高めで、ConoHa式にしては期間を伸ばしてもあまり安くならないが、時々80%キャッシュバックCPとかをやるので、それを使うと実質月100円弱とかで2GBプランが使えたりする。

これから登録すると俺に500円入ってくる感じです