2020/1/13 更新:最近終於把網站升級成 Vultr 的 High Frequency 方案(其實這個在2019年七八月已經推出了),真的很有感!! 主要是 CPU 升級跟硬碟改成 NVMe SSD,一個月也才多8美金,讀檔速度比之前快很多。
早在一年多前 Linode 的日本機房全數完售、沒法再增購或升級的那時候,我就開始使用很多人推薦的 Vultr VPS 主機。一直以來 Vultr 不論是主機的速度、機器與網路的穩定性,一直都有不錯的表現。
會用 Vultr 的 VPS 服務除了價格漂亮之外,最重要的是他有日本機房,,
Vultr 以小時計費,用多久、算多少錢。在後台就可以直接開設或關停主機,完全不需要等待人工確認或處理,要用就開、不用就關掉,很方便。
- 網站名稱:Vultr
- 網站網址:https://www.vultr.com
- Test IP、測試檔下載:按這裡
- 目前使用 Vultr 的網站:狂人論壇
Vultr 目前提供了美國、日本、新加坡、英國倫敦、法國巴黎、德國法蘭克福、荷蘭阿姆斯特丹、澳洲雪梨… 等橫跨歐、美、亞、澳洲.. 各地的機房可以選擇,每個主機也都提供了包含 Windows, Debian, Ubuntu, CentOS, Fedora, FreeBSD…等多種版本的作業系統可用, 幾乎新版作業系統推出沒多久後就有得用。如果沒有你要的版本,也可以自行上傳 ISO 檔,透過後台的 Console 來手動安裝、設定。
其實開始用 Vultr 之後才發現裡面有許多設計相當好用,譬如說前一陣子新增的防火牆功能,可以建立一組或多組IPv4 或 IPv6 的防火牆規則,然後讓帳號內的不同主機直接套用。據說這個免費的防火牆功能可以擋下小規模的 DDOS 攻擊,不過如果遇到比較大規模、嚴重一點攻擊,還有額外的 DDoS 防護方案可選擇,一個月只要 10 美金(有多大的能耐不知道,但至少有點用吧)。
Vultr 可支援 IPv6,不管 IPv4 或 IPv6 都能設定主機的 DNS 反查(一般人可能用不到,但看起來就是爽),每個主機都有詳細的圖表顯示當月的流量、頻寬、CPU、硬碟…等使用量,如果有什麼異常或資源不足等問題的話,可以先從 CPU Usage 跟 Disk Operations 等圖表看看有沒有什麼奇怪的數據。
另外,Vultr 有個免費的 Snapshot 功能,可以把目前設定好的主機狀態拍個 Snapshot 下來(全主機備份),等以後出問題或新增主機時,都能直接挖來用(新增主機的時候還可以直接套用先前的 Snapshot)。當然也有 Automatic backup 功能,備份起來更輕鬆方便,一個月也只要主機金額的 20%。
除了一般的 VPS,還有 Storage Instance 跟 Dedicated Instance 等服務,如果有一般小規模架站需求的話,Vultr 其實已經非常夠用且好用了,價格也相當划算。缺點的話…. 日本機房的硬體等級跟線路小輸 Linode 一點點,其他一切都好。
Vultr 網路連線、主機效能測試:
網路連線測試…
Google Chrome 瀏覽器下載測試…
FileZilla 透過 SFTP 的檔案下載測試…
BYTE UNIX Benchmarks 主機效能測試:(4CPU/8GB記憶體/4TB流量/每月$40美金的VPS)
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: TEST: GNU/Linux
OS: GNU/Linux — 4.9.0-3-amd64 — #1 SMP Debian 4.9.30-2+deb9u1 (2017-06-18)
Machine: x86_64 (unknown)
Language: en_US.utf8 (charmap=”UTF-8″, collate=”UTF-8″)
CPU 0: Virtual CPU a7769a6388d5 (4788.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: Virtual CPU a7769a6388d5 (4788.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 2: Virtual CPU a7769a6388d5 (4788.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 3: Virtual CPU a7769a6388d5 (4788.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
04:20:43 up 19 min, 1 user, load average: 0.00, 0.02, 0.04; runlevel 5
————————————————————————
Benchmark Run: Wed Jun 21 2017 04:20:43 – 04:48:47
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 26968817.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4532.3 MWIPS (9.9 s, 7 samples)
Execl Throughput 1917.6 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 847008.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 255226.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1661294.2 KBps (30.0 s, 2 samples)
Pipe Throughput 1789374.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 235705.3 lps (10.0 s, 7 samples)
Process Creation 5155.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4493.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1728.3 lpm (60.0 s, 2 samples)
System Call Overhead 2842470.8 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 26968817.4 2311.0
Double-Precision Whetstone 55.0 4532.3 824.1
Execl Throughput 43.0 1917.6 446.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 847008.5 2138.9
File Copy 256 bufsize 500 maxblocks 1655.0 255226.2 1542.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 1661294.2 2864.3
Pipe Throughput 12440.0 1789374.6 1438.4
Pipe-based Context Switching 4000.0 235705.3 589.3
Process Creation 126.0 5155.0 409.1
Shell Scripts (1 concurrent) 42.4 4493.5 1059.8
Shell Scripts (8 concurrent) 6.0 1728.3 2880.5
System Call Overhead 15000.0 2842470.8 1895.0
========
System Benchmarks Index Score 1260.6
————————————————————————
Benchmark Run: Wed Jun 21 2017 04:48:47 – 05:16:58
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 77032704.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 15695.4 MWIPS (9.8 s, 7 samples)
Execl Throughput 11160.3 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 810575.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 226494.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2278233.7 KBps (30.0 s, 2 samples)
Pipe Throughput 5357580.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching 727746.1 lps (10.0 s, 7 samples)
Process Creation 26561.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 21310.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 2554.9 lpm (60.1 s, 2 samples)
System Call Overhead 6215326.8 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 77032704.8 6600.9
Double-Precision Whetstone 55.0 15695.4 2853.7
Execl Throughput 43.0 11160.3 2595.4
File Copy 1024 bufsize 2000 maxblocks 3960.0 810575.1 2046.9
File Copy 256 bufsize 500 maxblocks 1655.0 226494.1 1368.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 2278233.7 3928.0
Pipe Throughput 12440.0 5357580.3 4306.7
Pipe-based Context Switching 4000.0 727746.1 1819.4
Process Creation 126.0 26561.2 2108.0
Shell Scripts (1 concurrent) 42.4 21310.8 5026.1
Shell Scripts (8 concurrent) 6.0 2554.9 4258.2
System Call Overhead 15000.0 6215326.8 4143.6
========
System Benchmarks Index Score 3100.1
======= Script description and score comparison completed! =======
第二次測試:
System Benchmarks Index Score 1324.2
System Benchmarks Index Score 3169.8
第三次測試:
System Benchmarks Index Score 1283.0
System Benchmarks Index Score 3169.2
CPU 效能:
Maximum prime number checked in CPU test: 10000
Test execution summary:
total time: 11.9097s
total number of events: 10000
total time taken by event execution: 11.9078
per-request statistics:
min: 1.11ms
avg: 1.19ms
max: 2.88ms
approx. 95 percentile: 1.47ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 11.9078/0.00
記憶體效能 Memory Reads:
Operations performed: 104857600 (2610613.59 ops/sec)
102400.00 MB transferred (2549.43 MB/sec)
Test execution summary:
total time: 40.1659s
total number of events: 104857600
total time taken by event execution: 31.9014
per-request statistics:
min: 0.00ms
avg: 0.00ms
max: 5.93ms
approx. 95 percentile: 0.00ms
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 31.9014/0.00
檔案存取效能 File I/O Benchmarks:
Operations performed: 104857600 (2716263.35 ops/sec)
102400.00 MB transferred (2652.60 MB/sec)
Test execution summary:
total time: 38.6036s
total number of events: 104857600
total time taken by event execution: 30.6521
per-request statistics:
min: 0.00ms
avg: 0.00ms
max: 4.87ms
approx. 95 percentile: 0.00ms
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 30.6521/0.00
root@briian:~# sysbench --test=fileio prepare
sysbench 0.4.12: multi-threaded system evaluation benchmark
128 files, 16384Kb each, 2048Mb total