大家可能知道微魔已经很久很久没有发VPS评测的帖子了(当然并不是没人找…),这次收到了Hostsailor官方的邀请,加上对于测评这一块其实一直想做一些特点出来,于是决定今天开始重新开始“VPS测评”板块。Hostsailor这家在2015年3月曾在微魔部落上出现(传送门),这家在2011年注册于迪拜(OMG,传说中全球有钱人的聚集地…),主营的服务器机房为欧洲的荷兰机房,资历算是不错。废话不多说,开始测评。
2017-7-20 更新:微魔在2017年7月中旬期间数次与该商家老板沟通,过程与结果都很不愉快(并非是分赃不均等龌龊的原因),因此撤回本文文末给予商家的优秀评价。
2017-7-20 Update: The last contact with Hostsailor made me so unconfortable. Both the boss and the supprot team showed different faces before and after. So I decided to revoke the “Editor’s Choice” award.
Hostsailor官版1G/40G/1T/Xen PV测评
近日,有不少朋友询问微魔发文量减少的原因,实际上在过去的一整年,微魔都在处于找工作的状态,虽然现在工作问题已经基本落实,但以后也不希望回到曾经每天1篇促销信息帖的日子了(好的促销依然会发),以后微魔会更关注VPS品质的评价和VPS一些应用类的教程,争取让每篇文章的质量有所提升,也希望大家能多提宝贵的意见。
测试机配置
- 1GB内存
- 40GB硬盘
- 1TB流量@1Gbps
- Xen PV/SolusVM
- 1个ipv4
- 无限ipv6
- Node:xen04
- 价格:5.59美元
7美元/月(使用终身八折优惠码:HS20OFFVPS)
一、软件部分
软件环境:Ubuntu 14.04 x86;值得一提的是,HostSailor的镜像中包括了OpenSUSE 13.1的镜像,在微魔的印象中提供这个Linux发行版的厂商非常之少
官方镜像安装完毕后,内存占用如下
~# free -m total used free shared buffers cached Mem: 999 322 676 0 29 244 -/+ buffers/cache: 48 950 Swap: 1023 0 1023
系统更新正常,没有出现Ubuntu常出现的软件源的问题(请别问我哪家有问题,解决方法参照《apt-get出现“no public key available…”的解决方法》)
Fetched 49.4 MB in 7s (6,328 kB/s)
二、硬件部分
1. CPU:4核CPU(E5620 @ 2.40GHz)
题外话,最近微魔在国外看到了一个争议贴,说检验CPU是可以看下model的输出,如果在Intel(R) Xeon(R) CPU前面有个QEMU,则性能上就要折损很多。
~# cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 44 model name : Intel(R) Xeon(R) CPU E5620 @ 2.40GHz stepping : 2 microcode : 0x14 cpu MHz : 2400.136 cache size : 12288 KB physical id : 0 siblings : 4 core id : 0 cpu cores : 1 apicid : 1 initial apicid : 1 fdiv_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu de tsc msr pae cx8 apic sep cmov pat clflush mmx fxsr sse s se2 ss ht nx constant_tsc pni pclmulqdq ssse3 sse4_1 sse4_2 popcnt aes hypervisor ida arat epb dtherm bogomips : 4800.27 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: processor : 1 vendor_id : GenuineIntel cpu family : 6 model : 44 model name : Intel(R) Xeon(R) CPU E5620 @ 2.40GHz stepping : 2 microcode : 0x14 cpu MHz : 2400.136 cache size : 12288 KB physical id : 0 siblings : 4 core id : 0 cpu cores : 1 apicid : 1 initial apicid : 1 fdiv_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu de tsc msr pae cx8 apic sep cmov pat clflush mmx fxsr sse s se2 ss ht nx constant_tsc pni pclmulqdq ssse3 sse4_1 sse4_2 popcnt aes hypervisor ida arat epb dtherm bogomips : 4800.27 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: processor : 2 vendor_id : GenuineIntel cpu family : 6 model : 44 model name : Intel(R) Xeon(R) CPU E5620 @ 2.40GHz stepping : 2 microcode : 0x14 cpu MHz : 2400.136 cache size : 12288 KB physical id : 0 siblings : 4 core id : 0 cpu cores : 1 apicid : 1 initial apicid : 1 fdiv_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu de tsc msr pae cx8 apic sep cmov pat clflush mmx fxsr sse s se2 ss ht nx constant_tsc pni pclmulqdq ssse3 sse4_1 sse4_2 popcnt aes hypervisor ida arat epb dtherm bogomips : 4800.27 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: processor : 3 vendor_id : GenuineIntel cpu family : 6 model : 44 model name : Intel(R) Xeon(R) CPU E5620 @ 2.40GHz stepping : 2 microcode : 0x14 cpu MHz : 2400.136 cache size : 12288 KB physical id : 0 siblings : 4 core id : 0 cpu cores : 1 apicid : 1 initial apicid : 1 fdiv_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu de tsc msr pae cx8 apic sep cmov pat clflush mmx fxsr sse s se2 ss ht nx constant_tsc pni pclmulqdq ssse3 sse4_1 sse4_2 popcnt aes hypervisor ida arat epb dtherm bogomips : 4800.27 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management:
2.网速测试:
~# wget http://cachefly.cachefly.net/100mb.test --2016-03-12 12:44:12-- http://cachefly.cachefly.net/100mb.test Resolving cachefly.cachefly.net (cachefly.cachefly.net)... 205.234.175.175 Connecting to cachefly.cachefly.net (cachefly.cachefly.net)|205.234.175.175|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: 100mb.test 100%[=============================================>] 104,857,600 46.3MB/s in 2.2s 2016-03-12 12:44:14 (46.3 MB/s) - 100mb.test saved [104857600/104857600]
各机房的速度
Atlanta的速度是令人发指的“异常”,考虑到各种意外在实际使用中也会发生,因此对此结果不再重新测试。
Speedtest (IPv4 only) Location Provider Speed CDN Cachefly 51.0MB/s Atlanta, GA, US Coloat 773KB/s Dallas, TX, US Softlayer 5.70MB/s Seattle, WA, US Softlayer 4.01MB/s San Jose, CA, US Softlayer 6.47MB/s Washington, DC, US Softlayer 2.10MB/s Tokyo, Japan Linode 7.92MB/s Singapore Softlayer 3.69MB/s Rotterdam, Netherlands id3.net 17.9MB/s Haarlem, Netherlands Leaseweb 50.6MB/s
国内外Ping值测速
作为欧洲的线路算是不错了
(2016-3-12 21:00测试)
(2016-3-13 14:30测试)
武汉电信Tracert
通过最多 30 个跃点跟踪到 185.82.*.* 的路由 1 <1 毫秒 <1 毫秒 <1 毫秒 59.71.255.254 2 3 ms 1 ms 2 ms 10.81.96.6 3 1 ms 1 ms 2 ms 58.48.110.225 4 4 ms 7 ms 5 ms 111.175.226.25 5 5 ms 6 ms 8 ms 111.175.210.221 6 21 ms 22 ms 18 ms 202.97.67.1 7 24 ms 23 ms 33 ms 202.97.33.194 8 181 ms * 22 ms 202.97.91.202 9 * 206 ms 208 ms 202.97.58.2 10 217 ms 220 ms 213 ms xe-10-2-3.lon25.ip4.gtt.net [46.33.88.61] 11 215 ms 215 ms 218 ms xe-7-1-3.ams12.ip4.gtt.net [89.149.180.133] 12 213 ms 213 ms 215 ms nforce-gw.ip4.gtt.net [46.33.91.70] 13 218 ms 217 ms 220 ms 109.201.142.235 14 245 ms 242 ms 302 ms 185.82.*.* 跟踪完成。
3.硬盘读写dd测试:270+M/s作为普通硬盘而言,可以说是表现很好。
~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 3.89213 s, 276 MB/s ~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 3.82058 s, 281 MB/s ~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 3.70624 s, 290 MB/s
4.Unix Bench跑分
在本文的Ubuntu 14.04下,测试时间为2016-3-13 19:10
结果如下
BYTE UNIX Benchmarks (Version 5.1.3) System: ****.hostsailor.com: GNU/Linux OS: GNU/Linux -- 3.13.0-37-generic -- #64-Ubuntu SMP Mon Sep 22 21:30:01 UTC 2014 Machine: i686 (i686) Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8") CPU 0: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (4800.3 bogomips) Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT CPU 1: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (4800.3 bogomips) Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT CPU 2: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (4800.3 bogomips) Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT CPU 3: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (4800.3 bogomips) Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT 09:17:45 up 12 min, 1 user, load average: 0.35, 0.35, 0.22; runlevel 2 ------------------------------------------------------------------------ Benchmark Run: Sun Mar 13 2016 09:17:45 - 09:46:03 4 CPUs in system; running 1 parallel copy of tests Dhrystone 2 using register variables 13596255.0 lps (10.0 s, 7 samples) Double-Precision Whetstone 2565.7 MWIPS (10.1 s, 7 samples) Execl Throughput 1015.7 lps (29.9 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 190216.2 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 67252.0 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 561476.9 KBps (30.0 s, 2 samples) Pipe Throughput 461999.1 lps (10.0 s, 7 samples) Pipe-based Context Switching 24886.1 lps (10.0 s, 7 samples) Process Creation 1975.3 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 3214.2 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 1060.9 lpm (60.0 s, 2 samples) System Call Overhead 573995.0 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 13596255.0 1165.1 Double-Precision Whetstone 55.0 2565.7 466.5 Execl Throughput 43.0 1015.7 236.2 File Copy 1024 bufsize 2000 maxblocks 3960.0 190216.2 480.3 File Copy 256 bufsize 500 maxblocks 1655.0 67252.0 406.4 File Copy 4096 bufsize 8000 maxblocks 5800.0 561476.9 968.1 Pipe Throughput 12440.0 461999.1 371.4 Pipe-based Context Switching 4000.0 24886.1 62.2 Process Creation 126.0 1975.3 156.8 Shell Scripts (1 concurrent) 42.4 3214.2 758.1 Shell Scripts (8 concurrent) 6.0 1060.9 1768.1 System Call Overhead 15000.0 573995.0 382.7 ======== System Benchmarks Index Score 434.3 ------------------------------------------------------------------------ Benchmark Run: Sun Mar 13 2016 09:46:03 - 10:14:23 4 CPUs in system; running 4 parallel copies of tests Dhrystone 2 using register variables 43878885.9 lps (10.0 s, 7 samples) Double-Precision Whetstone 9958.3 MWIPS (10.0 s, 7 samples) Execl Throughput 3610.3 lps (30.0 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 323593.3 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 56914.1 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 970488.2 KBps (30.0 s, 2 samples) Pipe Throughput 1728570.8 lps (10.0 s, 7 samples) Pipe-based Context Switching 171878.6 lps (10.0 s, 7 samples) Process Creation 6112.4 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 8263.0 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 1182.5 lpm (60.1 s, 2 samples) System Call Overhead 2017612.8 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 43878885.9 3760.0 Double-Precision Whetstone 55.0 9958.3 1810.6 Execl Throughput 43.0 3610.3 839.6 File Copy 1024 bufsize 2000 maxblocks 3960.0 323593.3 817.2 File Copy 256 bufsize 500 maxblocks 1655.0 56914.1 343.9 File Copy 4096 bufsize 8000 maxblocks 5800.0 970488.2 1673.3 Pipe Throughput 12440.0 1728570.8 1389.5 Pipe-based Context Switching 4000.0 171878.6 429.7 Process Creation 126.0 6112.4 485.1 Shell Scripts (1 concurrent) 42.4 8263.0 1948.8 Shell Scripts (8 concurrent) 6.0 1182.5 1970.9 System Call Overhead 15000.0 2017612.8 1345.1 ======== System Benchmarks Index Score 1123.0
微魔查看了一下网友上传的一些其他商家1G内存,HDD硬盘,10刀以下配置的套餐(下图),认为HostSailor的UnixBench得分(1123分)虽然称不上惊艳,但是也并不逊色(值得注意的是,似乎从性能发挥上来讲,传统观点认为OpenVZ构架的性能要比Xen好,但Xen胜在对超售的控制)。
5.在线率
待7天/15天后补充
2016-4-14更新:据原始测评31天后,更新最新结果如下。
总结
就本次的测试数据而言,HostSailor这款1G配置的Xen测试机的优势首先体现在性价比上,作为一款Xen构架的VPS,能够在1G内存配置上做到6美元以内(需结合优惠码)是非常难能可贵的,而且本文的测试也表明,其性能也是不错,配置上也不存在明显的短板;另外,荷兰的法律相对较为宽松,荷兰机房一直以来是做外贸朋友的首选之一,当然每个卖家会有自己的规定(查看HostSailor的TOS)。当然,荷兰机房被一般用户所诟病的主要是网速的问题,这也是“鱼与熊掌不可兼得”的问题。总之,从目前来看,微魔认为HostSailor的这款供微魔测试的Xen VPS在性价比上优势不错,最终评价结果微魔将在2周后统计了uptime之后截稿发布。
为期31天的测评结束,HostSailor测评机uptime时间为31天,在线率在本次测评中为100%,硬盘DD数据依旧维持在140M/s以上,根据以上测评结果,颁发微魔部落“编辑选择奖”(Editor’s Choice)奖章,商家引用参考如下,
“编辑选择奖”仅为微魔根据测评结果颁发,结果参考性能(CPU、硬盘、Benchmark等)及在线率(uptime)。测评结果仅对本次测评有效,非终身制。如对此次测评结果有任何异议,请在本文留言或给我发送邮件。
The Editor’s Choice Award from VMVPS is based on the results posted herein, which manily referred to the VPS performance, e.g., CPU model, HD and network speed and Benchmark test, as well as uptime data. Be aware, the title is not lifetime but valid for this test only. If you have any kind of objections, leave your comment here or send me an E-mail.
好长的测评文!看来是用心了
测评文好像也短不了…
这家貌似不错,准备买买看
目前看确实不错
我见过的逼格最高的测评文!
我收到过的最高逼格的评价!
Apologies from us/HostSailor for the unpleasant experience back in 2017 because of the failure in handling your case by our individual (already left) staff members. It has been very unfortunate for both you and us.
We do hope you can put that aside and move on for a win-win scenario ahead together with HostSailor. We believe that you will certainly enjoy collaborating with HostSailor in the future.
维魔部落的站长,你好!基于上次我们的邮件联络,本该早些在这里发回帖表达歉意的,由于一些事情和原因耽误了,到现在才来留言,非常抱歉。我们非常重视你在服务器测评方面的专业性,也非常遗憾几年前因为我们之前同事(已离职)的不善处理导致最终的不欢而散。
我们知道你很忙,但是我们诚心希望今后可以不计前嫌保留合作的可能性,至少不放弃一个可能双赢的机会,也许我们现在正在开发的Premium KVM VPS你会有兴趣了解 – 我会在产品开发完成以后再和你联系。
祝你一切安好顺利,贵站越来越受欢迎!
Glad to see your reply here. The story has passed for quite a long time. And it seems that you guys are starting a fresh start. That’s great. Hope you and your business going well in the future.
Thank you!