VPS推荐网HostKVM近期上线了香港湾仔机房的VPS服务器，基于KVM架构，特色就是带宽大，服务器接入在100Mbps的大带宽，相对于不少10Mbps-20Mbps左右的国内直连线路来说的确要大很多，不过线路不是CN2直连线路，ping波动还是非常大的，面向国内的带宽不会有很大，根据hostkvm长期运营的经验来看，不会接受长时间占满带宽的用户，所以还是比较适合建站的用户选择。下面是hostkvm湾仔机房常规套餐的简单测试，有喜欢的可以参考一下，了解一些CPU性能以及IO等，线路这个波动大。
1、基本信息
---------------------------------------------------------------------- CPU Model : QEMU Virtual CPU version 2.5+ CPU Cores : 2 Cores @ 2099.998 MHz x86_64 CPU Cache : 16384 KB OS : CentOS 6.6 (64 Bit) KVM Kernel : 2.6.32-504.el6.x86_64 Total Space : 20.0 GB (0.7 GB Used) Total RAM : 996 MB (263 MB Used 33 MB Buff) Total SWAP : 254 MB (0 MB Used) Uptime : 1 days 1 hour 31 min Load Average : 0.00, 0.00, 0.00 TCP CC : cubic ASN & ISP : AS9312, xTom Hong Kong Limited Organization : xTom Location : , / Region : ----------------------------------------------------------------------
CPU为2核，不过还是不能长时间占满，这个时间最好不要超过1小时；
2、硬盘IO读写测试
---------------------------------------------------------------------- I/O Speed( 1.0GB ) : 336 MB/s I/O Speed( 1.0GB ) : 342 MB/s I/O Speed( 1.0GB ) : 340 MB/s Average I/O Speed : 339.3 MB/s ----------------------------------------------------------------------
IO性能还是非常不错的，在hostkvm提供的VPS主机中处于优秀的位置，可以满足大部分的应用；
3、中国电信、联通、移动到VPS的上传、下载速度测试
---------------------------------------------------------------------- Node Name Upload Speed Download Speed Latency Speedtest.net 66.42 Mbit/s 76.77 Mbit/s 16.231 ms Fast.com 0.00 Mbit/s 76.5 Mbit/s - Nanjing CT 73.80 Mbit/s 73.41 Mbit/s - Hangzhou CT 68.12 Mbit/s 74.14 Mbit/s - Shanghai CU 75.59 Mbit/s 75.99 Mbit/s - Xi'an CU 69.05 Mbit/s 47.81 Mbit/s - Chongqing CU 76.61 Mbit/s 70.25 Mbit/s - Chengdu CM 53.47 Mbit/s 58.35 Mbit/s - Kunming CM 42.92 Mbit/s 71.28 Mbit/s - Guangzhou CM 43.71 Mbit/s 34.34 Mbit/s - -----------------------------------------xqblog-----------------------------
4、服务器全球节点文件下载速度测试
---------------------------------------------------------------------- Node Name IPv4 address Download Speed CacheFly 205.234.175.175 7.82MB/s Linode, Tokyo, JP 106.187.96.148 9.02MB/s Linode, Singapore, SG 139.162.23.4 9.05MB/s Linode, London, UK 176.58.107.39 7.21MB/s Linode, Frankfurt, DE 139.162.130.8 6.31MB/s Linode, Fremont, CA 50.116.14.9 7.94MB/s Softlayer, Dallas, TX 173.192.68.18 8.07MB/s Softlayer, Seattle, WA 67.228.112.250 7.59MB/s Softlayer, Frankfurt, DE 159.122.69.4 5.96MB/s Softlayer, Singapore, SG 119.81.28.170 9.09MB/s Softlayer, HongKong, CN 119.81.130.170 9.19MB/s -------------------------------------xqblog---------------------------------
目前处于空闲状态，所以各方面情况还是非常不错的，具体的反响需要长期的观测才行，所以一旦宿主主机处于饱和状态，稳定性影响不大，主要还是带宽问题；
5、UnixBench跑分测试
------------------------------------------------------------------------ Benchmark Run: Wed Nov 14 2018 14:09:20 - 14:38:29 2 CPUs in system; running 1 parallel copy of tests Dhrystone 2 using register variables 28538106.6 lps (10.0 s, 7 samples) Double-Precision Whetstone 3462.3 MWIPS (9.7 s, 7 samples) Execl Throughput 3639.1 lps (30.0 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 882151.6 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 237039.6 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 2625943.5 KBps (30.0 s, 2 samples) Pipe Throughput 1553657.5 lps (10.0 s, 7 samples) Pipe-based Context Switching 296490.1 lps (10.0 s, 7 samples) Process Creation 9390.5 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 5197.2 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 1148.5 lpm (60.0 s, 2 samples) System Call Overhead 2092573.7 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 28538106.6 2445.4 Double-Precision Whetstone 55.0 3462.3 629.5 Execl Throughput 43.0 3639.1 846.3 File Copy 1024 bufsize 2000 maxblocks 3960.0 882151.6 2227.7 File Copy 256 bufsize 500 maxblocks 1655.0 237039.6 1432.3 File Copy 4096 bufsize 8000 maxblocks 5800.0 2625943.5 4527.5 Pipe Throughput 12440.0 1553657.5 1248.9 Pipe-based Context Switching 4000.0 296490.1 741.2 Process Creation 126.0 9390.5 745.3 Shell Scripts (1 concurrent) 42.4 5197.2 1225.8 Shell Scripts (8 concurrent) 6.0 1148.5 1914.1 System Call Overhead 15000.0 2092573.7 1395.0 ======== System Benchmarks Index Score 1366.8 -------------------------------xqblog----------------------------------------- Benchmark Run: Wed Nov 14 14:09:20 - 14:38:29 2 CPUs in system; running 2 parallel copies of tests Dhrystone 2 using register variables 61212218.2 lps (10.0 s, 7 samples) Double-Precision Whetstone 7728.0 MWIPS (9.6 s, 7 samples) Execl Throughput 7480.1 lps (29.7 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 612073.8 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 182554.1 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 1834464.9 KBps (30.0 s, 2 samples) Pipe Throughput 3264773.0 lps (10.0 s, 7 samples) Pipe-based Context Switching 673449.1 lps (10.0 s, 7 samples) Process Creation 25968.2 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 9015.5 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 1333.1 lpm (60.0 s, 2 samples) System Call Overhead 3266168.2 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 61212218.2 5245.3 Double-Precision Whetstone 55.0 7728.0 1405.1 Execl Throughput 43.0 7480.1 1739.6 File Copy 1024 bufsize 2000 maxblocks 3960.0 612073.8 1545.6 File Copy 256 bufsize 500 maxblocks 1655.0 182554.1 1103.0 File Copy 4096 bufsize 8000 maxblocks 5800.0 1834464.9 3162.9 Pipe Throughput 12440.0 3264773.0 2624.4 Pipe-based Context Switching 4000.0 673449.1 1683.6 Process Creation 126.0 25968.2 2061.0 Shell Scripts (1 concurrent) 42.4 9015.5 2126.3 Shell Scripts (8 concurrent) 6.0 1333.1 2321.8 System Call Overhead 15000.0 3266168.2 2177.4 ======== System Benchmarks Index Score 2097.8
6、小结
在hostKVM旗下的所有VPS主机中处于不错的水平，可以满足大部分的建站任何，所以他家还是以建站位置；对于喜欢大带宽的用户是个不错的选择，再加上现在的优惠促销活动，还是蛮具有一定的性价比的。