最近感到Hostigation的网络越来越卡了,有时候开net speeder都没多大作用,于是又蠢蠢欲动准备搬迁。前些日子尝试了两个vps:lunanode和locvps。
lunanode是个相对较新的IDC,他们刚刚从芝加哥机房搬迁到多伦多机房,提供服务的方式类似于DigitalOcean,可以随时动态生成vps,按小时计费。不同的是lunanode是真正的KVM,而DO的KVM还没法自由更新内核。前段时间还推出了$7/month的2GB内存VPS,性价比算是挺高的了。
没有选择lunanode主要是因为两点:一是ping值太高,国内访问速度一般。到多伦多300-400ms的ping,看着实在太不爽;二是这个idc刚刚进入廉价vps市场不久,还没有证明其信誉的东西。在这个每个月都有几家idc出现几家idc跑路的时代,不得不考虑提供商信誉的问题。
尝试的另一家就是hostloc论坛的名人kwx开办的locvps,我选用的是其位于香港的机器。不得不说香港vps的响应速度实在是太让人舒服了,ssh操作如丝般顺滑,30多ms的ping,基本和国内服务器差不多,还不用纠结国内的备案。但是香港服务器也普遍有个硬伤:给的带宽太小太小!1M、2M的带宽,让人情何以堪啊。另外这家的vps跑分貌似也是不忍直视,附上晚上8、9点钟的UnixBench结果:
CPU 0: Intel(R) Xeon(R) CPU L5520 @ 2.27GHz (4533.6 bogomips) Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT CPU 1: Intel(R) Xeon(R) CPU L5520 @ 2.27GHz (4533.6 bogomips) Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT CPU 2: Intel(R) Xeon(R) CPU L5520 @ 2.27GHz (4533.6 bogomips) Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT CPU 3: Intel(R) Xeon(R) CPU L5520 @ 2.27GHz (4533.6 bogomips) Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT ------------------------------------------------------------------------ 4 CPUs in system; running 1 parallel copy of tests Dhrystone 2 using register variables 15194517.6 lps (10.0 s, 7 samples) Double-Precision Whetstone 2286.9 MWIPS (9.9 s, 7 samples) Execl Throughput 826.3 lps (30.0 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 72201.9 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 18439.9 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 255927.9 KBps (30.0 s, 2 samples) Pipe Throughput 82867.2 lps (10.0 s, 7 samples) Pipe-based Context Switching 23381.8 lps (10.0 s, 7 samples) Process Creation 1445.7 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 1866.0 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 763.6 lpm (60.1 s, 2 samples) System Call Overhead 529018.0 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 15194517.6 1302.0 Double-Precision Whetstone 55.0 2286.9 415.8 Execl Throughput 43.0 826.3 192.2 File Copy 1024 bufsize 2000 maxblocks 3960.0 72201.9 182.3 File Copy 256 bufsize 500 maxblocks 1655.0 18439.9 111.4 File Copy 4096 bufsize 8000 maxblocks 5800.0 255927.9 441.3 Pipe Throughput 12440.0 82867.2 66.6 Pipe-based Context Switching 4000.0 23381.8 58.5 Process Creation 126.0 1445.7 114.7 Shell Scripts (1 concurrent) 42.4 1866.0 440.1 Shell Scripts (8 concurrent) 6.0 763.6 1272.7 System Call Overhead 15000.0 529018.0 352.7 ======== System Benchmarks Index Score 256.8 ------------------------------------------------------------------------ 4 CPUs in system; running 4 parallel copies of tests Dhrystone 2 using register variables 55350933.6 lps (10.0 s, 7 samples) Double-Precision Whetstone 9114.3 MWIPS (10.4 s, 7 samples) Execl Throughput 3945.9 lps (30.0 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 93951.5 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 24577.1 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 361694.7 KBps (30.0 s, 2 samples) Pipe Throughput 172700.8 lps (10.0 s, 7 samples) Pipe-based Context Switching 72203.2 lps (10.0 s, 7 samples) Process Creation 6188.8 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 6281.9 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 839.8 lpm (60.2 s, 2 samples) System Call Overhead 1825540.3 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 55350933.6 4743.0 Double-Precision Whetstone 55.0 9114.3 1657.1 Execl Throughput 43.0 3945.9 917.6 File Copy 1024 bufsize 2000 maxblocks 3960.0 93951.5 237.3 File Copy 256 bufsize 500 maxblocks 1655.0 24577.1 148.5 File Copy 4096 bufsize 8000 maxblocks 5800.0 361694.7 623.6 Pipe Throughput 12440.0 172700.8 138.8 Pipe-based Context Switching 4000.0 72203.2 180.5 Process Creation 126.0 6188.8 491.2 Shell Scripts (1 concurrent) 42.4 6281.9 1481.6 Shell Scripts (8 concurrent) 6.0 839.8 1399.7 System Call Overhead 15000.0 1825540.3 1217.0 ======== System Benchmarks Index Score 642.2
话说分数低得有点不正常,毕竟我在上面跑过minecraft这种耗CPU的程序,并没有感觉到任何卡顿。
总结:如果你只是想买个vps用于个人翻墙,或者随便建个小站,那么香港vps是非常好的选择。不过即使是2M带宽,看720P Youtube还是有点吃力的。对我这种习惯美国1G带宽的人来说,2M小水管无法忍受,即使我的博客平时没多少PV。
也考虑过Linode,但是之前两次冲$5送$50的活动都错过了,总觉得没占到便宜心里不爽>_<其实是我的信用卡还没批下来
最后想起来还有个新加坡的OneAsiaHost,性价比还不错。走香港PCCW线路,到国内的ping在100以内。关键是带宽给得相对较大,有100M,能满足绝大部分需求了。于是搞了个,把博客搬到了上面。顺便也跑个分:
CPU 0: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (6784.9 bogomips) Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization ------------------------------------------------------------------------ 1 CPU in system; running 1 parallel copy of tests Dhrystone 2 using register variables 26391122.8 lps (10.0 s, 7 samples) Double-Precision Whetstone 3437.9 MWIPS (10.0 s, 7 samples) Execl Throughput 6179.7 lps (30.0 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 742768.3 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 202961.2 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 1887032.3 KBps (30.0 s, 2 samples) Pipe Throughput 1631065.4 lps (10.0 s, 7 samples) Pipe-based Context Switching 285466.2 lps (10.0 s, 7 samples) Process Creation 12007.3 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 4294.5 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 730.6 lpm (60.1 s, 2 samples) System Call Overhead 2162677.4 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 26391122.8 2261.5 Double-Precision Whetstone 55.0 3437.9 625.1 Execl Throughput 43.0 6179.7 1437.1 File Copy 1024 bufsize 2000 maxblocks 3960.0 742768.3 1875.7 File Copy 256 bufsize 500 maxblocks 1655.0 202961.2 1226.4 File Copy 4096 bufsize 8000 maxblocks 5800.0 1887032.3 3253.5 Pipe Throughput 12440.0 1631065.4 1311.1 Pipe-based Context Switching 4000.0 285466.2 713.7 Process Creation 126.0 12007.3 953.0 Shell Scripts (1 concurrent) 42.4 4294.5 1012.9 Shell Scripts (8 concurrent) 6.0 730.6 1217.6 System Call Overhead 15000.0 2162677.4 1441.8 ======== System Benchmarks Index Score 1303.7
虽然不是服务器CPU,跑分结果还是很漂亮的。
本文为悠然居(https://wordpress.youran.me/)的原创文章,转载请注明出处!
求问跑分程序…
UnixBench:
https://code.google.com/p/byte-unixbench/downloads/list
突然发现你这儿机器好快…
主要是后台开启了WP Total Cache,配合memcached,效果不错。
好久没用mac os了,突然发现mac下的chrome显示我博客title的那三个字有问题,每个字最右边就像被竖直地切掉一小段似的,Firefox/Safari就没这个问题。你那里显示如何?
我这儿显示似乎正常,大约是字体问题?