0x01 前言
kdatacenter是韩国的一家VPS服务商，因为地理位置的优势，所以网络速度较为理想。可能是因为韩国地少，宽带贵的原因，他家的VPS非常昂贵，完全是土豪级别的。
之前一直想做个测试，无奈最低配置也需要每月19美元，摸摸自己的荷包还是不敢下手。有位网友得知此情况，特意借用他的VPS给我做个简单的性能测试，在这里要感谢@Gary。
为感谢@，以下是Gary在该厂商的推荐计划地址，如果大家有需要，可点击以下地址进行购买:
他家的VPS分为LIGHT与STANDARD两个版本，以下是两个版本的价格：
从价目表中可以看出价格较其他厂商高出不少，配置看上去也一般，而且还限流量，看上去性价比确实不高。
0x02 配置
这次测试的VPS配置是最低的一级，信息如下:
- CPU：1核心
- 内存：1G
- 硬盘：100G SSD
- 带宽：1G
- 流量：500G
- 系统：centos7
以上配置需要月付19美元，使用KVM虚拟化平台，具体的CPU型号则无法得知。本次测试机的IP地址在以下C段中：
- 27.102.115.0/24
该IP地址属于韩国的一家电信企业：GNJ-AS-KR DAOU TECHNOLOGY, KR。信息不多，也没什么可以关注的点，接下来直接进行测试。
0x03 网络测试
0x03.1 ping
首先是长ping测试，以下是测试结果：
--- 27.102.115.xxx ping statistics --- 6402 packets transmitted, 6218 received, 2% packet loss, time 6406035ms rtt min/avg/max/mdev = 66.834/111.271/306.081/46.722 ms
共计ping了6402次，有2%的丢包，而且平均延迟在111ms。
我进行这个测试是在北京时间晚上的8点，测试时发现一个有趣的现象：在ping了约45分钟后延迟从135ms左右下降到75ms左右，不知道是不是因为网络压力下降的原因导致的。
0x03.2 wget
随后安装个nginx，使用浏览器通过HTTP协议下载1G大小的文件，以下是结果：
[root@web ~]# wget http://27.102.115.xxx/1gb-1.test --2018-10-30 20:44:57-- http://27.102.115.xxx/1gb-1.test Connecting to 27.102.115.xxx:80... connected. HTTP request sent, awaiting response... 200 OK Length: 1073741824 (1.0G) [application/octet-stream] Saving to: ‘1gb-1.test’ 100%[===================================>] 1,073,741,824 3.50MB/s in 4m 32s 2018-10-30 20:49:32 (3.76 MB/s) - ‘1gb-1.test’ saved [1073741824/1073741824]
我的位置在广东深圳，使用电信游戏宽带，下行200M。使用wget下载文件的速度非常稳定，保持在3.5m/s，这速度还是不错的，至少比众多美欧和日本的厂家要好。
0x03.3 路由追踪
以下是路由追踪的结果：
从路由追踪的结果可以看出，大陆的数据基本是在北京上岸，而香港则绕路日本。如果主要用户是在东北，则延迟可能会很低，而且速度也较快；如果用户在南方，则延迟较高。
因为没有其他国家访客访问的需求，所以没有测试其他国家的链路情况。
0x03.4 speedtest
下面是speedtest的结果：
# 测试点为韩国kdatacenter Retrieving speedtest.net configuration... Testing from Daou Technology (27.102.115.xxx)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by kdatacenter.com (Seoul) [8.34 km]: 4.295 ms Testing download speed...... Download: 303.70 Mbit/s Testing upload speed............................ # 测试点为辽宁联通 [root@localhost ~]# speedtest --server 5017 Retrieving speedtest.net configuration... Testing from Daou Technology (27.102.115.xxx)... Retrieving speedtest.net server list... Retrieving information for the selected server... Hosted by China Unicom,Liaoning Branch (Shenyang) [538.87 km]: 138.721 ms Testing download speed...... Download: 62.18 Mbit/s Testing upload speed............................ Upload: 110.89 Mbit/s # 测试点为上海联通 [root@localhost ~]# speedtest --server 5083 Retrieving speedtest.net configuration... Testing from Daou Technology (27.102.115.xxx)... Retrieving speedtest.net server list... Retrieving information for the selected server... Hosted by Shanghai Branch, China Unicom (Shanghai) [862.74 km]: 28.568 ms Testing download speed...... Download: 178.87 Mbit/s Testing upload speed............................ Upload: 291.04 Mbit/s # 测试点为浙江电信 [root@localhost ~]# speedtest --server 7509 Retrieving speedtest.net configuration... Testing from Daou Technology (27.102.115.xxx)... Retrieving speedtest.net server list... Retrieving information for the selected server... Hosted by China Telecom ZheJiang Branch (Hangzhou) [1022.46 km]: 170.07 ms Testing download speed...... Download: 8.94 Mbit/s Testing upload speed............................ Upload: 47.75 Mbit/s # 测试点为广州电信 [root@localhost ~]# speedtest --server 17251 Retrieving speedtest.net configuration... Testing from Daou Technology (27.102.115.xxx)... Retrieving speedtest.net server list... Retrieving information for the selected server... Hosted by ChinaTelecom-GZ (Guangzhou) [2065.53 km]: 146.181 ms Testing download speed...... Download: 13.21 Mbit/s Testing upload speed............................ Upload: 76.50 Mbit/s # 测试点为海口移动 [root@localhost ~]# speedtest --server 16503 Retrieving speedtest.net configuration... Testing from Daou Technology (27.102.115.xxx)... Retrieving speedtest.net server list... Retrieving information for the selected server... Hosted by Chinamobile,Hainan (Haikou) [2521.38 km]: 165.776 ms Testing download speed...... Download: 4.19 Mbit/s Testing upload speed............................ Upload: 13.75 Mbit/s # 测试点为广州移动 [root@localhost ~]# speedtest --server 6611 Retrieving speedtest.net configuration... Testing from Daou Technology (27.102.115.xxx)... Retrieving speedtest.net server list... Retrieving information for the selected server... Hosted by China Mobile,Guangdong (Guangzhou) [2065.53 km]: 228.918 ms Testing download speed...... Download: 47.26 Mbit/s Testing upload speed............................ Upload: 19.55 Mbit/s
联通的速度较快，其次是移动和电信；而且北方的速度较南方要快。
0x04 性能测试
0x04.1 dd磁盘测试
[root@localhost codex]# dd bs=1M count=1024 if=/dev/zero of=1gb-1.test conv=fdatasync 1024+0 records in 1024+0 records out 1073741824 bytes (1.1 GB) copied, 1.78542 s, 601 MB/s [root@localhost codex]# dd bs=1M count=1024 if=/dev/zero of=1gb-2.test conv=fdatasync 1024+0 records in 1024+0 records out 1073741824 bytes (1.1 GB) copied, 1.49546 s, 718 MB/s [root@localhost codex]# dd bs=1M count=1024 if=/dev/zero of=1gb-3.test conv=fdatasync 1024+0 records in 1024+0 records out 1073741824 bytes (1.1 GB) copied, 1.52634 s, 703 MB/s [root@localhost codex]# dd bs=1M count=1024 if=/dev/zero of=1gb-4.test conv=fdatasync 1024+0 records in 1024+0 records out 1073741824 bytes (1.1 GB) copied, 1.471 s, 730 MB/s [root@localhost codex]# dd bs=1M count=1024 if=/dev/zero of=1gb-5.test conv=fdatasync 1024+0 records in 1024+0 records out 1073741824 bytes (1.1 GB) copied, 1.18966 s, 903 MB/s
磁盘速度非常好，而且磁盘空间有100G，毕竟是SSD，那么大的容量，也是要花不少钱的。
0x04.2 fio
CRC 加密测试：
[root@localhost ~]# fio -crctest md5: 428.75 MiB/sec crc64: 332.50 MiB/sec crc32: 293.47 MiB/sec crc32c: 328.80 MiB/sec crc16: 318.03 MiB/sec crc7: 358.97 MiB/sec sha1: 309.79 MiB/sec sha256: 145.47 MiB/sec sha512: 238.34 MiB/sec xxhash: 4462.50 MiB/sec murmur3: 2343.36 MiB/sec jhash: 1651.57 MiB/sec fnv: 2730.93 MiB/sec sha3-224: 99.80 MiB/sec sha3-256: 99.05 MiB/sec sha3-384: 71.86 MiB/sec sha3-512: 50.45 MiB/sec
通过与之前测试的VPS结果进行比对，kdatacenter的性能十分理想，要知道这只是单核心的VPS。
4k随机读写测试：
[root@localhost ~]# fio -filename=/root/bench/test.fio -direct=1 -rw=rw -bs=4k -size 1G -numjobs=8 -runtime=30 -group_reporting -name=file
file: (g=0): rw=rw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=psync, iodepth=1
...
fio-3.11-30-g19971
Starting 8 processes
Jobs: 8 (f=8): [M(8)][100.0%][r=11.6MiB/s,w=11.9MiB/s][r=2977,w=3059 IOPS][eta 00m:00s]
file: (groupid=0, jobs=8): err= 0: pid=25422: Tue Oct 30 11:12:14 2018
read: IOPS=3785, BW=14.8MiB/s (15.5MB/s)(444MiB/30004msec)
clat (usec): min=120, max=123943, avg=1303.55, stdev=2173.49
lat (usec): min=121, max=123944, avg=1304.67, stdev=2173.50
clat percentiles (usec):
| 1.00th=[ 176], 5.00th=[ 371], 10.00th=[ 478], 20.00th=[ 701],
| 30.00th=[ 840], 40.00th=[ 979], 50.00th=[ 1123], 60.00th=[ 1254],
| 70.00th=[ 1434], 80.00th=[ 1647], 90.00th=[ 1991], 95.00th=[ 2343],
| 99.00th=[ 3621], 99.50th=[ 6128], 99.90th=[38011], 99.95th=[54264],
| 99.99th=[74974]
bw ( KiB/s): min= 832, max= 2560, per=12.49%, avg=1891.74, stdev=279.37, samples=480
iops : min= 208, max= 640, avg=472.89, stdev=69.83, samples=480
write: IOPS=3786, BW=14.8MiB/s (15.5MB/s)(444MiB/30004msec)
clat (usec): min=97, max=123664, avg=798.30, stdev=1867.28
lat (usec): min=98, max=123665, avg=799.52, stdev=1867.33
clat percentiles (usec):
| 1.00th=[ 114], 5.00th=[ 117], 10.00th=[ 120], 20.00th=[ 127],
| 30.00th=[ 147], 40.00th=[ 163], 50.00th=[ 326], 60.00th=[ 807],
| 70.00th=[ 1090], 80.00th=[ 1369], 90.00th=[ 1762], 95.00th=[ 2147],
| 99.00th=[ 3130], 99.50th=[ 4080], 99.90th=[21103], 99.95th=[41157],
| 99.99th=[74974]
bw ( KiB/s): min= 880, max= 2800, per=12.49%, avg=1892.46, stdev=292.47, samples=480
iops : min= 220, max= 700, avg=473.07, stdev=73.10, samples=480
lat (usec) : 100=0.01%, 250=25.90%, 500=5.80%, 750=8.94%, 1000=13.48%
lat (msec) : 2=37.67%, 4=7.54%, 10=0.39%, 20=0.13%, 50=0.10%
lat (msec) : 100=0.04%, 250=0.01%
cpu : usr=0.66%, sys=1.93%, ctx=449404, majf=0, minf=95
IO depths : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=113580,113620,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: bw=14.8MiB/s (15.5MB/s), 14.8MiB/s-14.8MiB/s (15.5MB/s-15.5MB/s), io=444MiB (465MB), run=30004-30004msec
WRITE: bw=14.8MiB/s (15.5MB/s), 14.8MiB/s-14.8MiB/s (15.5MB/s-15.5MB/s), io=444MiB (465MB), run=30004-30004msec
Disk stats (read/write):
vda: ios=113168/113370, merge=0/607, ticks=23586/15827, in_queue=47449, util=99.67%
32k随机读写测试：
[root@localhost ~]# fio -filename=/root/bench/test.fio -direct=1 -rw=rw -bs=32k -size 1G -numjobs=8 -runtime=30 -group_reporting -name=file
file: (g=0): rw=rw, bs=(R) 32.0KiB-32.0KiB, (W) 32.0KiB-32.0KiB, (T) 32.0KiB-32.0KiB, ioengine=psync, iodepth=1
...
fio-3.11-30-g19971
Starting 8 processes
Jobs: 8 (f=8): [M(8)][100.0%][r=58.0MiB/s,w=58.7MiB/s][r=1887,w=1878 IOPS][eta 00m:00s]
file: (groupid=0, jobs=8): err= 0: pid=25481: Tue Oct 30 11:13:19 2018
read: IOPS=2179, BW=68.1MiB/s (71.4MB/s)(2043MiB/30005msec)
clat (usec): min=220, max=111893, avg=2308.69, stdev=3027.44
lat (usec): min=221, max=111894, avg=2309.87, stdev=3027.50
clat percentiles (usec):
| 1.00th=[ 285], 5.00th=[ 685], 10.00th=[ 914], 20.00th=[ 1221],
| 30.00th=[ 1450], 40.00th=[ 1663], 50.00th=[ 1893], 60.00th=[ 2147],
| 70.00th=[ 2474], 80.00th=[ 2868], 90.00th=[ 3621], 95.00th=[ 4424],
| 99.00th=[ 8225], 99.50th=[15795], 99.90th=[51643], 99.95th=[60556],
| 99.99th=[92799]
bw ( KiB/s): min= 4864, max=12416, per=12.49%, avg=8712.82, stdev=1197.11, samples=480
iops : min= 152, max= 388, avg=272.21, stdev=37.43, samples=480
write: IOPS=2181, BW=68.2MiB/s (71.5MB/s)(2045MiB/30005msec)
clat (usec): min=150, max=103894, avg=1348.34, stdev=2678.27
lat (usec): min=152, max=103896, avg=1350.25, stdev=2678.32
clat percentiles (usec):
| 1.00th=[ 172], 5.00th=[ 176], 10.00th=[ 180], 20.00th=[ 188],
| 30.00th=[ 202], 40.00th=[ 225], 50.00th=[ 652], 60.00th=[ 1385],
| 70.00th=[ 1795], 80.00th=[ 2278], 90.00th=[ 2933], 95.00th=[ 3621],
| 99.00th=[ 6652], 99.50th=[10945], 99.90th=[40633], 99.95th=[57410],
| 99.99th=[92799]
bw ( KiB/s): min= 4800, max=13504, per=12.49%, avg=8720.60, stdev=1343.92, samples=480
iops : min= 150, max= 422, avg=272.45, stdev=42.01, samples=480
lat (usec) : 250=22.69%, 500=3.14%, 750=2.48%, 1000=4.39%
lat (msec) : 2=31.78%, 4=30.20%, 10=4.63%, 20=0.38%, 50=0.21%
lat (msec) : 100=0.09%, 250=0.01%
cpu : usr=0.43%, sys=1.30%, ctx=258804, majf=0, minf=110
IO depths : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=65391,65442,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: bw=68.1MiB/s (71.4MB/s), 68.1MiB/s-68.1MiB/s (71.4MB/s-71.4MB/s), io=2043MiB (2143MB), run=30005-30005msec
WRITE: bw=68.2MiB/s (71.5MB/s), 68.2MiB/s-68.2MiB/s (71.5MB/s-71.5MB/s), io=2045MiB (2144MB), run=30005-30005msec
Disk stats (read/write):
vda: ios=65120/65227, merge=0/113, ticks=34655/11807, in_queue=52778, util=99.65%
1m随机读写测试：
[root@localhost ~]# fio -filename=/root/bench/test.fio -direct=1 -rw=rw -bs=1m -size 1G -numjobs=8 -runtime=30 -group_reporting -name=file
file: (g=0): rw=rw, bs=(R) 1024KiB-1024KiB, (W) 1024KiB-1024KiB, (T) 1024KiB-1024KiB, ioengine=psync, iodepth=1
...
fio-3.11-30-g19971
Starting 8 processes
Jobs: 3 (f=3): [M(1),_(1),M(1),_(3),M(1),_(1)][100.0%][r=128MiB/s,w=142MiB/s][r=128,w=142 IOPS][eta 00m:00s]
file: (groupid=0, jobs=8): err= 0: pid=25539: Tue Oct 30 11:14:14 2018
read: IOPS=167, BW=167MiB/s (175MB/s)(4012MiB/24004msec)
clat (msec): min=2, max=120, avg=30.06, stdev=12.87
lat (msec): min=2, max=120, avg=30.06, stdev=12.87
clat percentiles (msec):
| 1.00th=[ 4], 5.00th=[ 15], 10.00th=[ 18], 20.00th=[ 21],
| 30.00th=[ 24], 40.00th=[ 26], 50.00th=[ 29], 60.00th=[ 32],
| 70.00th=[ 35], 80.00th=[ 39], 90.00th=[ 45], 95.00th=[ 52],
| 99.00th=[ 75], 99.50th=[ 94], 99.90th=[ 117], 99.95th=[ 121],
| 99.99th=[ 122]
bw ( KiB/s): min= 8192, max=34746, per=12.55%, avg=21486.04, stdev=4746.56, samples=375
iops : min= 8, max= 33, avg=20.87, stdev= 4.66, samples=375
write: IOPS=174, BW=174MiB/s (183MB/s)(4180MiB/24004msec)
clat (msec): min=2, max=127, avg=16.48, stdev=16.53
lat (msec): min=2, max=127, avg=16.52, stdev=16.54
clat percentiles (msec):
| 1.00th=[ 3], 5.00th=[ 3], 10.00th=[ 3], 20.00th=[ 3],
| 30.00th=[ 3], 40.00th=[ 3], 50.00th=[ 6], 60.00th=[ 22],
| 70.00th=[ 27], 80.00th=[ 32], 90.00th=[ 37], 95.00th=[ 44],
| 99.00th=[ 60], 99.50th=[ 80], 99.90th=[ 110], 99.95th=[ 118],
| 99.99th=[ 128]
bw ( KiB/s): min= 6131, max=51200, per=12.60%, avg=22468.86, stdev=7356.00, samples=375
iops : min= 5, max= 50, avg=21.83, stdev= 7.20, samples=375
lat (msec) : 4=25.96%, 10=1.18%, 20=10.41%, 50=58.45%, 100=3.67%
lat (msec) : 250=0.32%
cpu : usr=0.14%, sys=0.23%, ctx=16222, majf=0, minf=88
IO depths : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=4012,4180,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: bw=167MiB/s (175MB/s), 167MiB/s-167MiB/s (175MB/s-175MB/s), io=4012MiB (4207MB), run=24004-24004msec
WRITE: bw=174MiB/s (183MB/s), 174MiB/s-174MiB/s (183MB/s-183MB/s), io=4180MiB (4383MB), run=24004-24004msec
Disk stats (read/write):
vda: ios=3991/4223, merge=0/112, ticks=37738/9764, in_queue=48575, util=99.15%
毕竟是SSD，读写性能都很优异。
0x04.3 ramsmp
[root@localhost ramsmp-3.5.0]# ./ramsmp -b 1 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes INTEGER & WRITING 1 Kb block: 23788.11 MB/s INTEGER & WRITING 2 Kb block: 20391.37 MB/s INTEGER & WRITING 4 Kb block: 24102.00 MB/s INTEGER & WRITING 8 Kb block: 22444.77 MB/s INTEGER & WRITING 16 Kb block: 23860.52 MB/s INTEGER & WRITING 32 Kb block: 19736.45 MB/s INTEGER & WRITING 64 Kb block: 20858.28 MB/s INTEGER & WRITING 128 Kb block: 21266.16 MB/s INTEGER & WRITING 256 Kb block: 18582.32 MB/s INTEGER & WRITING 512 Kb block: 18921.95 MB/s INTEGER & WRITING 1024 Kb block: 18566.75 MB/s INTEGER & WRITING 2048 Kb block: 18077.74 MB/s INTEGER & WRITING 4096 Kb block: 15721.72 MB/s INTEGER & WRITING 8192 Kb block: 16391.93 MB/s INTEGER & WRITING 16384 Kb block: 10579.26 MB/s INTEGER & WRITING 32768 Kb block: 8188.77 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 2 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes INTEGER & READING 1 Kb block: 51135.81 MB/s INTEGER & READING 2 Kb block: 51727.74 MB/s INTEGER & READING 4 Kb block: 48676.73 MB/s INTEGER & READING 8 Kb block: 51781.53 MB/s INTEGER & READING 16 Kb block: 37974.66 MB/s INTEGER & READING 32 Kb block: 46202.94 MB/s INTEGER & READING 64 Kb block: 32603.21 MB/s INTEGER & READING 128 Kb block: 33716.76 MB/s INTEGER & READING 256 Kb block: 29645.10 MB/s INTEGER & READING 512 Kb block: 23635.02 MB/s INTEGER & READING 1024 Kb block: 19958.15 MB/s INTEGER & READING 2048 Kb block: 22328.28 MB/s INTEGER & READING 4096 Kb block: 20784.17 MB/s INTEGER & READING 8192 Kb block: 18004.33 MB/s INTEGER & READING 16384 Kb block: 12606.12 MB/s INTEGER & READING 32768 Kb block: 7490.63 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 3 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes INTEGER Copy: 9928.34 MB/s INTEGER Scale: 9735.97 MB/s INTEGER Add: 10460.51 MB/s INTEGER Triad: 10420.26 MB/s --- INTEGER AVERAGE: 10136.27 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 4 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes FL-POINT & WRITING 1 Kb block: 24733.65 MB/s FL-POINT & WRITING 2 Kb block: 25263.19 MB/s FL-POINT & WRITING 4 Kb block: 25188.47 MB/s FL-POINT & WRITING 8 Kb block: 21291.99 MB/s FL-POINT & WRITING 16 Kb block: 24800.86 MB/s FL-POINT & WRITING 32 Kb block: 23660.57 MB/s FL-POINT & WRITING 64 Kb block: 21675.25 MB/s FL-POINT & WRITING 128 Kb block: 19090.47 MB/s FL-POINT & WRITING 256 Kb block: 19539.12 MB/s FL-POINT & WRITING 512 Kb block: 18368.39 MB/s FL-POINT & WRITING 1024 Kb block: 17413.08 MB/s FL-POINT & WRITING 2048 Kb block: 16316.37 MB/s FL-POINT & WRITING 4096 Kb block: 17751.33 MB/s FL-POINT & WRITING 8192 Kb block: 15709.42 MB/s FL-POINT & WRITING 16384 Kb block: 11316.55 MB/s FL-POINT & WRITING 32768 Kb block: 7699.53 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 5 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes FL-POINT & READING 1 Kb block: 25444.93 MB/s FL-POINT & READING 2 Kb block: 25498.99 MB/s FL-POINT & READING 4 Kb block: 24645.90 MB/s FL-POINT & READING 8 Kb block: 21219.33 MB/s FL-POINT & READING 16 Kb block: 25136.78 MB/s FL-POINT & READING 32 Kb block: 24976.09 MB/s FL-POINT & READING 64 Kb block: 24825.83 MB/s FL-POINT & READING 128 Kb block: 24762.51 MB/s FL-POINT & READING 256 Kb block: 20836.66 MB/s FL-POINT & READING 512 Kb block: 22658.14 MB/s FL-POINT & READING 1024 Kb block: 22764.14 MB/s FL-POINT & READING 2048 Kb block: 20524.23 MB/s FL-POINT & READING 4096 Kb block: 21124.76 MB/s FL-POINT & READING 8192 Kb block: 20438.62 MB/s FL-POINT & READING 16384 Kb block: 12177.15 MB/s FL-POINT & READING 32768 Kb block: 9136.85 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 6 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes FL-POINT Copy: 6887.73 MB/s FL-POINT Scale: 7144.93 MB/s FL-POINT Add: 8583.59 MB/s FL-POINT Triad: 9182.52 MB/s --- FL-POINT AVERAGE: 7949.69 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 7 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes MMX & WRITING 1 Kb block: 25711.73 MB/s MMX & WRITING 2 Kb block: 24823.14 MB/s MMX & WRITING 4 Kb block: 25296.15 MB/s MMX & WRITING 8 Kb block: 22146.58 MB/s MMX & WRITING 16 Kb block: 24918.37 MB/s MMX & WRITING 32 Kb block: 24668.23 MB/s MMX & WRITING 64 Kb block: 21996.34 MB/s MMX & WRITING 128 Kb block: 19348.85 MB/s MMX & WRITING 256 Kb block: 20863.07 MB/s MMX & WRITING 512 Kb block: 18856.07 MB/s MMX & WRITING 1024 Kb block: 16405.12 MB/s MMX & WRITING 2048 Kb block: 18356.39 MB/s MMX & WRITING 4096 Kb block: 18109.22 MB/s MMX & WRITING 8192 Kb block: 15042.59 MB/s MMX & WRITING 16384 Kb block: 11406.13 MB/s MMX & WRITING 32768 Kb block: 7222.79 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 8 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes MMX & READING 1 Kb block: 51813.86 MB/s MMX & READING 2 Kb block: 51715.98 MB/s MMX & READING 4 Kb block: 52176.18 MB/s MMX & READING 8 Kb block: 52337.52 MB/s MMX & READING 16 Kb block: 50996.55 MB/s MMX & READING 32 Kb block: 39057.76 MB/s MMX & READING 64 Kb block: 51068.40 MB/s MMX & READING 128 Kb block: 50760.51 MB/s MMX & READING 256 Kb block: 50897.74 MB/s MMX & READING 512 Kb block: 50728.28 MB/s MMX & READING 1024 Kb block: 49847.58 MB/s MMX & READING 2048 Kb block: 51054.89 MB/s MMX & READING 4096 Kb block: 31915.82 MB/s MMX & READING 8192 Kb block: 22908.34 MB/s MMX & READING 16384 Kb block: 25454.98 MB/s MMX & READING 32768 Kb block: 24858.95 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 9 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes MMX Copy: 10089.34 MB/s MMX Scale: 10671.88 MB/s MMX Add: 11147.19 MB/s MMX Triad: 11757.71 MB/s --- MMX AVERAGE: 10916.53 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 10 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes SSE & WRITING 1 Kb block: 49261.70 MB/s SSE & WRITING 2 Kb block: 49150.71 MB/s SSE & WRITING 4 Kb block: 50502.59 MB/s SSE & WRITING 8 Kb block: 50959.79 MB/s SSE & WRITING 16 Kb block: 50300.46 MB/s SSE & WRITING 32 Kb block: 50700.14 MB/s SSE & WRITING 64 Kb block: 30106.32 MB/s SSE & WRITING 128 Kb block: 24667.06 MB/s SSE & WRITING 256 Kb block: 27589.03 MB/s SSE & WRITING 512 Kb block: 22739.01 MB/s SSE & WRITING 1024 Kb block: 21675.35 MB/s SSE & WRITING 2048 Kb block: 18641.41 MB/s SSE & WRITING 4096 Kb block: 20975.57 MB/s SSE & WRITING 8192 Kb block: 20052.84 MB/s SSE & WRITING 16384 Kb block: 11174.76 MB/s SSE & WRITING 32768 Kb block: 8394.22 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 11 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes SSE & READING 1 Kb block: 103137.49 MB/s SSE & READING 2 Kb block: 104808.34 MB/s SSE & READING 4 Kb block: 104377.57 MB/s SSE & READING 8 Kb block: 104447.37 MB/s SSE & READING 16 Kb block: 104400.72 MB/s SSE & READING 32 Kb block: 105826.47 MB/s SSE & READING 64 Kb block: 102381.79 MB/s SSE & READING 128 Kb block: 99526.80 MB/s SSE & READING 256 Kb block: 105257.84 MB/s SSE & READING 512 Kb block: 100977.87 MB/s SSE & READING 1024 Kb block: 97424.41 MB/s SSE & READING 2048 Kb block: 98216.42 MB/s SSE & READING 4096 Kb block: 34526.05 MB/s SSE & READING 8192 Kb block: 34880.29 MB/s SSE & READING 16384 Kb block: 31465.31 MB/s SSE & READING 32768 Kb block: 30170.77 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 12 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes SSE Copy: 9377.42 MB/s SSE Scale: 10265.84 MB/s SSE Add: 11492.49 MB/s SSE Triad: 11404.29 MB/s --- SSE AVERAGE: 10635.01 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 13 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes MMX (nt) & WRITING 1 Kb block: 5341.18 MB/s MMX (nt) & WRITING 2 Kb block: 5671.43 MB/s MMX (nt) & WRITING 4 Kb block: 5356.38 MB/s MMX (nt) & WRITING 8 Kb block: 5456.84 MB/s MMX (nt) & WRITING 16 Kb block: 5476.65 MB/s MMX (nt) & WRITING 32 Kb block: 5599.62 MB/s MMX (nt) & WRITING 64 Kb block: 5423.02 MB/s MMX (nt) & WRITING 128 Kb block: 5597.12 MB/s MMX (nt) & WRITING 256 Kb block: 5415.85 MB/s MMX (nt) & WRITING 512 Kb block: 5496.82 MB/s MMX (nt) & WRITING 1024 Kb block: 5714.30 MB/s MMX (nt) & WRITING 2048 Kb block: 5497.90 MB/s MMX (nt) & WRITING 4096 Kb block: 5494.16 MB/s MMX (nt) & WRITING 8192 Kb block: 5415.15 MB/s MMX (nt) & WRITING 16384 Kb block: 5191.40 MB/s MMX (nt) & WRITING 32768 Kb block: 5401.95 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 14 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes MMX (nt) & READING 1 Kb block: 48228.80 MB/s MMX (nt) & READING 2 Kb block: 45994.45 MB/s MMX (nt) & READING 4 Kb block: 45883.46 MB/s MMX (nt) & READING 8 Kb block: 36294.30 MB/s MMX (nt) & READING 16 Kb block: 46466.99 MB/s MMX (nt) & READING 32 Kb block: 45812.80 MB/s MMX (nt) & READING 64 Kb block: 45483.20 MB/s MMX (nt) & READING 128 Kb block: 43808.32 MB/s MMX (nt) & READING 256 Kb block: 44615.91 MB/s MMX (nt) & READING 512 Kb block: 45108.15 MB/s MMX (nt) & READING 1024 Kb block: 45361.61 MB/s MMX (nt) & READING 2048 Kb block: 34828.18 MB/s MMX (nt) & READING 4096 Kb block: 35383.20 MB/s MMX (nt) & READING 8192 Kb block: 32253.52 MB/s MMX (nt) & READING 16384 Kb block: 30773.38 MB/s MMX (nt) & READING 32768 Kb block: 30829.57 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 15 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes MMX (nt) Copy: 6179.55 MB/s [NTA prefetch] MMX (nt) Scale: 6164.91 MB/s [NTA prefetch] MMX (nt) Add: 7897.33 MB/s [T0 prefetch] MMX (nt) Triad: 7223.90 MB/s [T0 prefetch] --- MMX (nt) AVERAGE: 6866.42 MB/s ---------------------------------------------------------------------------------- [root@localhost ramsmp-3.5.0]# ./ramsmp -b 16 RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes SSE & WRITING (nt) 1 Kb block: 5610.24 MB/s SSE & WRITING (nt) 2 Kb block: 5668.65 MB/s SSE & WRITING (nt) 4 Kb block: 5615.00 MB/s SSE & WRITING (nt) 8 Kb block: 5553.18 MB/s SSE & WRITING (nt) 16 Kb block: 5627.70 MB/s SSE & WRITING (nt) 32 Kb block: 5647.62 MB/s SSE & WRITING (nt) 64 Kb block: 5666.26 MB/s SSE & WRITING (nt) 128 Kb block: 5584.79 MB/s SSE & WRITING (nt) 256 Kb block: 5816.26 MB/s SSE & WRITING (nt) 512 Kb block: 5760.47 MB/s SSE & WRITING (nt) 1024 Kb block: 5663.83 MB/s SSE & WRITING (nt) 2048 Kb block: 5429.08 MB/s SSE & WRITING (nt) 4096 Kb block: 5567.91 MB/s SSE & WRITING (nt) 8192 Kb block: 5820.33 MB/s SSE & WRITING (nt) 16384 Kb block: 5294.21 MB/s SSE & WRITING (nt) 32768 Kb block: 5668.03 MB/s
内存性能较为优异。
0x04.4 Unix Benchmark
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: localhost.localdomain: GNU/Linux
OS: GNU/Linux -- 4.17.11-1.el7.elrepo.x86_64 -- #1 SMP Sat Jul 28 09:33:09 EDT 2018
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Common KVM processor (5200.0 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
11:27:56 up 89 days, 21:12, 2 users, load average: 1.11, 1.56, 1.18; runlevel 2018-08-01
------------------------------------------------------------------------
Benchmark Run: Tue Oct 30 2018 11:27:56 - 11:56:02
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 35078475.0 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4145.9 MWIPS (10.0 s, 7 samples)
Execl Throughput 3794.2 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 398719.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 105408.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1123975.6 KBps (30.0 s, 2 samples)
Pipe Throughput 616305.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching 143260.0 lps (10.0 s, 7 samples)
Process Creation 9038.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 5143.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 699.8 lpm (60.1 s, 2 samples)
System Call Overhead 487009.1 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 35078475.0 3005.9
Double-Precision Whetstone 55.0 4145.9 753.8
Execl Throughput 43.0 3794.2 882.4
File Copy 1024 bufsize 2000 maxblocks 3960.0 398719.7 1006.9
File Copy 256 bufsize 500 maxblocks 1655.0 105408.0 636.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 1123975.6 1937.9
Pipe Throughput 12440.0 616305.7 495.4
Pipe-based Context Switching 4000.0 143260.0 358.1
Process Creation 126.0 9038.7 717.4
Shell Scripts (1 concurrent) 42.4 5143.9 1213.2
Shell Scripts (8 concurrent) 6.0 699.8 1166.4
System Call Overhead 15000.0 487009.1 324.7
========
System Benchmarks Index Score 851.5
单核心得到851分，分数适中。
0x05 结语
kdatacenter性能中规中矩，对得起这个价格。估计是因为韩国带宽价格过高，才会拉升了售价，但售价太高，只能用磁盘空间来弥补了。但SSD的性能良好，做个企业站点还是没问题的，甚至做小论坛也绰绰有余。
至于网络则建议北方的朋友使用，做站点为南方或其他地区的客户提供服务，则建议套一层CDN。
kdatacenter除了价格贵还限流量，其他都很好，资金允许的情况下可以优先考虑。
