0x01 前言
最近发现krypt有优惠活动，所以我购买了一个独立服务器进行测试。因为购买购买的是配置最低的机器，所以性能并不是很好，另外网络也不太适合中国大陆的用户。
首先krypt是一家老牌的独立主机服务商，网络和机器都非常好，有专门为中国大陆准备的高级网络选项，网络直连大陆电信、联通和移动。
在租用独立服务器达到标准（一般为一年）后，该服务器的所有权归租用者所有。届时可以将服务器转移到其他服务商进行托管，也可以放在krypt机房继续托管，还可以邮寄回国内。
我还发现krypt现在也有云服务器了，近一个月有促销活动，价格低廉而且内存翻倍，续费价格也不会增加。
我所购买的服务器位于美国本土，网络虽然是直连中国大陆，但速度依旧不容乐观。云服务器部分有美国东西海岸和新加坡数据中心，新加坡数据中心的速度估计会好点。但因为费用的关系我没有购买测试，如果你愿意给予我提供资助，请通过本页面的捐赠方式进行捐赠，非常感谢。
0x02 配置
我的服务器于今天13点购买的，18点40分收到配置完成的邮件。因为我是刚接触krypt，所以用户级别比较低，如果级别高的经销商或用户购买新的独立服务器，配置时间会大大减少。
以下是服务器的配置信息：
- CPU：Outlet – Atom D510 1.66GHz
- 内存：2GB RAM
- 硬盘：500GB SATA HDD
- IP地址：5个 IPv4
- 网络：1000Mbps
- 流量：5TB
- 其他：中国优化网络
如果没有选择“中国优化网络”的话，流量为10TB，而且价格会少2美金左右。以上配置的月付费用为：38.45美金。以下为配置及费用截图：
通过查询，发现Intel Atom® Processor D510这个CPU为2010第一季度的产品，为2和4线程，且仅支持DDR2 800/667 内存。
通过这个信息可以发现，这个配置的服务器其实就是一台好几年前的台式机，性能肯定不足以和称为服务器的硬件相匹配。不过对这个价格来说，是超值的。
年付还有打折，折算成人民币大概为2800元，加上网络，绝对比很多云服务商的价格要低。
0x03 网络测试
0x03.1 ping
首先是ping测试，以下是结果：
--- ip1.us.t1.bever.in ping statistics --- 5413 packets transmitted, 4480 received, 17% packet loss, time 5415752ms rtt min/avg/max/mdev = 153.833/169.204/325.330/24.787 ms
使用我家里的网络（深圳电信，百兆光纤）长时间ping位于美国的这台独立服务器，平均170ms的延迟，可以接受，毕竟跨越了一片大海，但是高达17%的丢包率却是个大问题。不知道位于美国的其他机房会不会这样情况。
0x03.2 wget
然后是使用wget下载文件，首先是香港节点：
[root@hk1 ~]$ wget http://ip1.us.t1.bever.in/1gb-1.test --2017-07-02 00:38:55-- http://ip1.us.t1.bever.in/1gb-1.test Resolving ip1.us.t1.bever.in (ip1.us.t1.bever.in)... 98.126.129.178 Connecting to ip1.us.t1.bever.in (ip1.us.t1.bever.in)|98.126.129.178|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 1073741824 (1.0G) [application/octet-stream] Saving to: ‘1gb-1.test’ 1% [==> ] 14,188,015 359KB/s eta 44m 45s
想不到使用香港的网络也那么慢，只有500k/s上下。以下是我家里的网络：
[root@misc ~]# wget http://ip1.us.t1.bever.in/1gb-1.test --2017-07-02 00:38:00-- http://ip1.us.t1.bever.in/1gb-1.test Resolving ip1.us.t1.bever.in (ip1.us.t1.bever.in)... 98.126.129.178 Connecting to ip1.us.t1.bever.in (ip1.us.t1.bever.in)|98.126.129.178|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 1073741824 (1.0G) [application/octet-stream] Saving to: ‘1gb-1.test’ 0% [ ] 3,900,695 19.6KB/s eta 15h 6m
速度更慢，只有20k/s上下。
0x03.3 路由追踪
以下是路由追踪结果：
从路由追踪中可以确认是直连国内的，只是线路不太好，速度比较慢。虽然延迟可以接受，但不是使用CN2线路。
0x03.4 speedtest
以下是speedtest的测试结果：
#天津联通 Retrieving speedtest.net configuration... Testing from Krypt (98.126.129.178)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by China Unicom (Tianjin) [10126.07 km]: 213.398 ms Testing download speed................................................................................ Download: 20.61 Mbit/s Testing upload speed................................................................................................ Upload: 3.61 Mbit/s #青岛联通 Retrieving speedtest.net configuration... Testing from Krypt (98.126.129.178)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by China Unicom (Qingdao) [10176.15 km]: 184.762 ms Testing download speed................................................................................ Download: 16.19 Mbit/s Testing upload speed................................................................................................ Upload: 58.34 Mbit/s #上海电信 Retrieving speedtest.net configuration... Testing from Krypt (98.126.129.178)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by China Telecom (Shanghai) [10483.06 km]: 255.029 ms Testing download speed................................................................................ Download: 18.05 Mbit/s Testing upload speed................................................................................................ Upload: 8.26 Mbit/s #浙江移动 Retrieving speedtest.net configuration... Testing from Krypt (98.126.129.178)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by China Mobile Group Zhejiang Branch (Hangzhou) [10647.97 km]: 302.686 ms Testing download speed................................................................................ Download: 57.43 Mbit/s Testing upload speed................................................................................................ Upload: 20.60 Mbit/s #成都电子科技大学 Retrieving speedtest.net configuration... Testing from Krypt (98.126.129.178)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by University of Electronic Science and Technology of China (Chengdu) [11607.55 km]: 429.184 ms Testing download speed................................................................................ Download: 11.89 Mbit/s Testing upload speed................................................................................................ Upload: 7.28 Mbit/s #深圳移动 Retrieving speedtest.net configuration... Testing from Krypt (98.126.129.178)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by China Mobile Group GuangDong (Shenzhen) [11682.79 km]: 231.645 ms Testing download speed................................................................................ Download: 16.76 Mbit/s Testing upload speed................................................................................................ Upload: 4.89 Mbit/s #香港 Retrieving speedtest.net configuration... Testing from Krypt (98.126.129.178)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by Interoute VDC (Hong Kong) [11703.62 km]: 211.628 ms Testing download speed................................................................................ Download: 26.54 Mbit/s Testing upload speed................................................................................................ Upload: 69.14 Mbit/s #昆明联通 Retrieving speedtest.net configuration... Testing from Krypt (98.126.129.178)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by Yunnan Chinaunicom (Kunming) [12195.15 km]: 549.172 ms Testing download speed................................................................................ Download: 16.49 Mbit/s Testing upload speed................................................................................................ Upload: 61.96 Mbit/s #昆明电信 Retrieving speedtest.net configuration... Testing from Krypt (98.126.129.178)... Retrieving speedtest.net server list... Selecting best server based on ping... Hosted by China Telecom (Kunming) [12195.15 km]: 934.485 ms Testing download speed................................................................................ Download: 19.84 Mbit/s Testing upload speed................................................................................................ Upload: 2.17 Mbit/s
联通的速度最快，其次是移动，电信最慢，5m/s以内。
0x04 性能测试
0x04.1 dd磁盘测试
[root@AT166-7318 temp]# 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, 14.0982 s, 76.2 MB/s [root@AT166-7318 temp]# 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, 14.57 s, 73.7 MB/s [root@AT166-7318 temp]# 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, 15.013 s, 71.5 MB/s [root@AT166-7318 temp]# 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, 11.1174 s, 96.6 MB/s [root@AT166-7318 temp]# 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, 15.6958 s, 68.4 MB/s
因为是独立服务器，没有做RAID，而且使用的是普通SATA硬盘，所以磁盘速度惨不忍睹。
0x04.2 fio
CRC 加密测试：
[root@AT166-7318 ~]# fio -crctest md5: 220.19 MB/sec crc64: 155.45 MB/sec crc32: 129.58 MB/sec crc32c: 155.41 MB/sec crc16: 129.56 MB/sec crc7: 141.44 MB/sec sha1: 98.45 MB/sec sha256: 36.12 MB/sec sha512: 65.55 MB/sec xxhash: 514.70 MB/sec murmur3: 340.45 MB/sec jhash: 663.93 MB/sec fnv: 664.52 MB/sec
4k随机读写测试：
[root@AT166-7318 ~]# 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=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=1
...
fio-2.2.8
Starting 8 processes
file: Laying out IO file(s) (1 file(s) / 1024MB)
Jobs: 8 (f=8): [M(8)] [100.0% done] [4123KB/4347KB/0KB /s] [1030/1086/0 iops] [eta 00m:00s]
file: (groupid=0, jobs=8): err= 0: pid=14449: Sat Jul 1 06:59:42 2017
read : io=88372KB, bw=2944.8KB/s, iops=736, runt= 30010msec
clat (usec): min=164, max=789482, avg=5800.99, stdev=29654.82
lat (usec): min=166, max=789484, avg=5803.73, stdev=29654.84
clat percentiles (usec):
| 1.00th=[ 189], 5.00th=[ 237], 10.00th=[ 241], 20.00th=[ 251],
| 30.00th=[ 270], 40.00th=[ 394], 50.00th=[ 548], 60.00th=[ 628],
| 70.00th=[ 700], 80.00th=[ 1128], 90.00th=[ 7648], 95.00th=[15936],
| 99.00th=[168960], 99.50th=[224256], 99.90th=[325632], 99.95th=[362496],
| 99.99th=[749568]
bw (KB /s): min= 3, max= 3664, per=13.96%, avg=411.02, stdev=608.84
write: io=86096KB, bw=2868.1KB/s, iops=717, runt= 30010msec
clat (usec): min=175, max=789265, avg=5167.00, stdev=29754.86
lat (usec): min=178, max=789268, avg=5170.57, stdev=29754.88
clat percentiles (usec):
| 1.00th=[ 274], 5.00th=[ 286], 10.00th=[ 290], 20.00th=[ 298],
| 30.00th=[ 306], 40.00th=[ 378], 50.00th=[ 564], 60.00th=[ 644],
| 70.00th=[ 700], 80.00th=[ 788], 90.00th=[ 1592], 95.00th=[ 8512],
| 99.00th=[168960], 99.50th=[216064], 99.90th=[337920], 99.95th=[415744],
| 99.99th=[733184]
bw (KB /s): min= 4, max= 3760, per=14.03%, avg=402.26, stdev=607.39
lat (usec) : 250=9.80%, 500=37.76%, 750=27.80%, 1000=6.18%
lat (msec) : 2=6.02%, 4=0.88%, 10=6.45%, 20=1.94%, 50=0.79%
lat (msec) : 100=0.60%, 250=1.42%, 500=0.32%, 750=0.02%, 1000=0.01%
cpu : usr=0.31%, sys=2.25%, ctx=43678, majf=0, minf=269
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 : total=r=22093/w=21524/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: io=88372KB, aggrb=2944KB/s, minb=2944KB/s, maxb=2944KB/s, mint=30010msec, maxt=30010msec
WRITE: io=86096KB, aggrb=2868KB/s, minb=2868KB/s, maxb=2868KB/s, mint=30010msec, maxt=30010msec
Disk stats (read/write):
dm-1: ios=22027/21416, merge=0/0, ticks=126877/107930, in_queue=235503, util=99.84%, aggrios=21971/21399, aggrmerge=51/58, aggrticks=117936/102388, aggrin_queue=220366, aggrutil=99.78%
sda: ios=21971/21399, merge=51/58, ticks=117936/102388, in_queue=220366, util=99.78%
32k随机读写测试：
[root@AT166-7318 ~]# 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=32K-32K/32K-32K/32K-32K, ioengine=sync, iodepth=1
...
fio-2.2.8
Starting 8 processes
Jobs: 8 (f=8): [M(8)] [100.0% done] [22259KB/21301KB/0KB /s] [695/665/0 iops] [eta 00m:00s]
file: (groupid=0, jobs=8): err= 0: pid=14489: Sat Jul 1 07:01:10 2017
read : io=608960KB, bw=20289KB/s, iops=634, runt= 30015msec
clat (usec): min=277, max=7367.7K, avg=7740.92, stdev=143671.28
lat (usec): min=280, max=7367.7K, avg=7743.62, stdev=143671.29
clat percentiles (usec):
| 1.00th=[ 628], 5.00th=[ 844], 10.00th=[ 884], 20.00th=[ 908],
| 30.00th=[ 948], 40.00th=[ 980], 50.00th=[ 1048], 60.00th=[ 1128],
| 70.00th=[ 1256], 80.00th=[ 2024], 90.00th=[ 7520], 95.00th=[10944],
| 99.00th=[40704], 99.50th=[84480], 99.90th=[1548288], 99.95th=[5144576],
| 99.99th=[7110656]
bw (KB /s): min= 10, max=12135, per=23.85%, avg=4839.52, stdev=3290.24
write: io=594464KB, bw=19806KB/s, iops=618, runt= 30015msec
clat (usec): min=397, max=6041.7K, avg=4954.32, stdev=101157.27
lat (usec): min=405, max=6041.7K, avg=4964.01, stdev=101157.30
clat percentiles (usec):
| 1.00th=[ 540], 5.00th=[ 892], 10.00th=[ 916], 20.00th=[ 940],
| 30.00th=[ 964], 40.00th=[ 996], 50.00th=[ 1020], 60.00th=[ 1080],
| 70.00th=[ 1176], 80.00th=[ 1384], 90.00th=[ 2800], 95.00th=[ 7648],
| 99.00th=[24704], 99.50th=[50944], 99.90th=[444416], 99.95th=[1449984],
| 99.99th=[6062080]
bw (KB /s): min= 5, max=11648, per=23.76%, avg=4705.67, stdev=3257.39
lat (usec) : 500=0.85%, 750=0.29%, 1000=41.42%
lat (msec) : 2=40.56%, 4=4.73%, 10=7.74%, 20=2.66%, 50=1.09%
lat (msec) : 100=0.29%, 250=0.21%, 500=0.05%, 750=0.02%, 1000=0.01%
lat (msec) : 2000=0.04%, >=2000=0.05%
cpu : usr=0.53%, sys=2.08%, ctx=37635, majf=0, minf=268
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 : total=r=19030/w=18577/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: io=608960KB, aggrb=20288KB/s, minb=20288KB/s, maxb=20288KB/s, mint=30015msec, maxt=30015msec
WRITE: io=594464KB, aggrb=19805KB/s, minb=19805KB/s, maxb=19805KB/s, mint=30015msec, maxt=30015msec
Disk stats (read/write):
dm-1: ios=18993/18551, merge=0/0, ticks=129336/99814, in_queue=318316, util=99.84%, aggrios=18963/18519, aggrmerge=15/10, aggrticks=111832/80513, aggrin_queue=261656, aggrutil=99.84%
sda: ios=18963/18519, merge=15/10, ticks=111832/80513, in_queue=261656, util=99.84%
1m随机读写测试：
[root@AT166-7318 ~]# 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=1M-1M/1M-1M/1M-1M, ioengine=sync, iodepth=1
...
fio-2.2.8
Starting 8 processes
Jobs: 8 (f=8): [M(8)] [100.0% done] [39896KB/40919KB/0KB /s] [38/39/0 iops] [eta 00m:00s]
file: (groupid=0, jobs=8): err= 0: pid=14513: Sat Jul 1 07:02:46 2017
read : io=1348.0MB, bw=45917KB/s, iops=44, runt= 30062msec
clat (msec): min=4, max=9218, avg=108.50, stdev=591.02
lat (msec): min=4, max=9218, avg=108.51, stdev=591.02
clat percentiles (msec):
| 1.00th=[ 5], 5.00th=[ 8], 10.00th=[ 9], 20.00th=[ 10],
| 30.00th=[ 12], 40.00th=[ 19], 50.00th=[ 25], 60.00th=[ 32],
| 70.00th=[ 40], 80.00th=[ 50], 90.00th=[ 75], 95.00th=[ 128],
| 99.00th=[ 2409], 99.50th=[ 6259], 99.90th=[ 7111], 99.95th=[ 9241],
| 99.99th=[ 9241]
bw (KB /s): min= 145, max=39009, per=29.08%, avg=13350.81, stdev=9781.55
write: io=1155.0MB, bw=39343KB/s, iops=38, runt= 30062msec
clat (msec): min=4, max=10595, avg=81.03, stdev=533.91
lat (msec): min=4, max=10595, avg=81.31, stdev=533.91
clat percentiles (msec):
| 1.00th=[ 5], 5.00th=[ 6], 10.00th=[ 9], 20.00th=[ 10],
| 30.00th=[ 10], 40.00th=[ 10], 50.00th=[ 10], 60.00th=[ 11],
| 70.00th=[ 18], 80.00th=[ 24], 90.00th=[ 47], 95.00th=[ 81],
| 99.00th=[ 2409], 99.50th=[ 3752], 99.90th=[ 7111], 99.95th=[10552],
| 99.99th=[10552]
bw (KB /s): min= 144, max=33158, per=30.67%, avg=12065.08, stdev=9000.14
lat (msec) : 10=40.99%, 20=16.34%, 50=28.25%, 100=9.03%, 250=1.52%
lat (msec) : 500=1.04%, 750=1.20%, 1000=0.20%, 2000=0.12%, >=2000=1.32%
cpu : usr=0.19%, sys=0.31%, ctx=2530, majf=0, minf=270
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 : total=r=1348/w=1155/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: io=1348.0MB, aggrb=45916KB/s, minb=45916KB/s, maxb=45916KB/s, mint=30062msec, maxt=30062msec
WRITE: io=1155.0MB, aggrb=39342KB/s, minb=39342KB/s, maxb=39342KB/s, mint=30062msec, maxt=30062msec
Disk stats (read/write):
dm-1: ios=2690/2319, merge=0/0, ticks=250762/257701, in_queue=655067, util=99.80%, aggrios=2696/2322, aggrmerge=0/3, aggrticks=271116/333436, aggrin_queue=604527, aggrutil=99.80%
sda: ios=2696/2322, merge=0/3, ticks=271116/333436, in_queue=604527, util=99.80%
从上面的测试结果可以看出服务器的磁盘性能不太好，我们家中的硬盘读写速度一般有100m/s左右。不知道是这台服务器用了很久还是其他原因，读写速度和IO都不太理想。
0x04.3 ramsmp
紧接着是内存测试，以下是测试结果：
[root@AT166-7318 ramsmp-3.5.0]# sh start_test.sh 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: 12965.33 MB/s INTEGER & WRITING 2 Kb block: 13017.93 MB/s INTEGER & WRITING 4 Kb block: 13040.11 MB/s INTEGER & WRITING 8 Kb block: 13044.63 MB/s INTEGER & WRITING 16 Kb block: 6803.52 MB/s INTEGER & WRITING 32 Kb block: 6799.75 MB/s INTEGER & WRITING 64 Kb block: 6805.28 MB/s INTEGER & WRITING 128 Kb block: 6774.62 MB/s INTEGER & WRITING 256 Kb block: 4961.24 MB/s INTEGER & WRITING 512 Kb block: 2055.06 MB/s INTEGER & WRITING 1024 Kb block: 1723.13 MB/s INTEGER & WRITING 2048 Kb block: 1916.41 MB/s INTEGER & WRITING 4096 Kb block: 1877.80 MB/s INTEGER & WRITING 8192 Kb block: 1882.88 MB/s INTEGER & WRITING 16384 Kb block: 1883.89 MB/s INTEGER & WRITING 32768 Kb block: 1841.15 MB/s 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: 25319.23 MB/s INTEGER & READING 2 Kb block: 25523.92 MB/s INTEGER & READING 4 Kb block: 25611.09 MB/s INTEGER & READING 8 Kb block: 25664.96 MB/s INTEGER & READING 16 Kb block: 25492.56 MB/s INTEGER & READING 32 Kb block: 10054.30 MB/s INTEGER & READING 64 Kb block: 10012.51 MB/s INTEGER & READING 128 Kb block: 10018.01 MB/s INTEGER & READING 256 Kb block: 9623.09 MB/s INTEGER & READING 512 Kb block: 7560.01 MB/s INTEGER & READING 1024 Kb block: 5104.76 MB/s INTEGER & READING 2048 Kb block: 4583.09 MB/s INTEGER & READING 4096 Kb block: 4562.77 MB/s INTEGER & READING 8192 Kb block: 4574.75 MB/s INTEGER & READING 16384 Kb block: 4578.17 MB/s INTEGER & READING 32768 Kb block: 4583.95 MB/s RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes INTEGER Copy: 2469.11 MB/s INTEGER Scale: 2359.53 MB/s INTEGER Add: 2314.88 MB/s INTEGER Triad: 2368.78 MB/s --- INTEGER AVERAGE: 2378.07 MB/s 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: 4343.96 MB/s FL-POINT & WRITING 2 Kb block: 4334.32 MB/s FL-POINT & WRITING 4 Kb block: 4354.26 MB/s FL-POINT & WRITING 8 Kb block: 4360.68 MB/s FL-POINT & WRITING 16 Kb block: 4360.32 MB/s FL-POINT & WRITING 32 Kb block: 4368.72 MB/s FL-POINT & WRITING 64 Kb block: 4371.72 MB/s FL-POINT & WRITING 128 Kb block: 4370.52 MB/s FL-POINT & WRITING 256 Kb block: 4364.51 MB/s FL-POINT & WRITING 512 Kb block: 3987.07 MB/s FL-POINT & WRITING 1024 Kb block: 2350.50 MB/s FL-POINT & WRITING 2048 Kb block: 1898.88 MB/s FL-POINT & WRITING 4096 Kb block: 1882.08 MB/s FL-POINT & WRITING 8192 Kb block: 1884.94 MB/s FL-POINT & WRITING 16384 Kb block: 1885.49 MB/s FL-POINT & WRITING 32768 Kb block: 1883.06 MB/s 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: 6542.95 MB/s FL-POINT & READING 2 Kb block: 6487.14 MB/s FL-POINT & READING 4 Kb block: 6553.49 MB/s FL-POINT & READING 8 Kb block: 6567.23 MB/s FL-POINT & READING 16 Kb block: 6566.80 MB/s FL-POINT & READING 32 Kb block: 6569.04 MB/s FL-POINT & READING 64 Kb block: 6571.85 MB/s FL-POINT & READING 128 Kb block: 6569.96 MB/s FL-POINT & READING 256 Kb block: 6454.07 MB/s FL-POINT & READING 512 Kb block: 5894.93 MB/s FL-POINT & READING 1024 Kb block: 4973.44 MB/s FL-POINT & READING 2048 Kb block: 4695.45 MB/s FL-POINT & READING 4096 Kb block: 4627.13 MB/s FL-POINT & READING 8192 Kb block: 4588.88 MB/s FL-POINT & READING 16384 Kb block: 4655.32 MB/s FL-POINT & READING 32768 Kb block: 4648.80 MB/s 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: 2475.06 MB/s FL-POINT Scale: 2476.27 MB/s FL-POINT Add: 2718.63 MB/s FL-POINT Triad: 2727.84 MB/s --- FL-POINT AVERAGE: 2599.45 MB/s 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: 24763.61 MB/s MMX & WRITING 2 Kb block: 24953.35 MB/s MMX & WRITING 4 Kb block: 25048.43 MB/s MMX & WRITING 8 Kb block: 24428.09 MB/s MMX & WRITING 16 Kb block: 24700.28 MB/s MMX & WRITING 32 Kb block: 11892.62 MB/s MMX & WRITING 64 Kb block: 11984.37 MB/s MMX & WRITING 128 Kb block: 11907.88 MB/s MMX & WRITING 256 Kb block: 11925.04 MB/s MMX & WRITING 512 Kb block: 6142.44 MB/s MMX & WRITING 1024 Kb block: 2339.74 MB/s MMX & WRITING 2048 Kb block: 1916.94 MB/s MMX & WRITING 4096 Kb block: 1880.02 MB/s MMX & WRITING 8192 Kb block: 1884.92 MB/s MMX & WRITING 16384 Kb block: 1886.33 MB/s MMX & WRITING 32768 Kb block: 1883.78 MB/s 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: 24765.04 MB/s MMX & READING 2 Kb block: 24953.97 MB/s MMX & READING 4 Kb block: 25050.66 MB/s MMX & READING 8 Kb block: 24769.51 MB/s MMX & READING 16 Kb block: 24929.14 MB/s MMX & READING 32 Kb block: 25001.97 MB/s MMX & READING 64 Kb block: 25019.93 MB/s MMX & READING 128 Kb block: 24789.81 MB/s MMX & READING 256 Kb block: 24778.99 MB/s MMX & READING 512 Kb block: 24087.72 MB/s MMX & READING 1024 Kb block: 24084.62 MB/s MMX & READING 2048 Kb block: 24077.14 MB/s MMX & READING 4096 Kb block: 8166.91 MB/s MMX & READING 8192 Kb block: 5980.08 MB/s MMX & READING 16384 Kb block: 5238.29 MB/s MMX & READING 32768 Kb block: 4928.76 MB/s RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes MMX Copy: 2378.10 MB/s MMX Scale: 2380.71 MB/s MMX Add: 2338.56 MB/s MMX Triad: 2356.01 MB/s --- MMX AVERAGE: 2363.35 MB/s 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: 44325.59 MB/s SSE & WRITING 2 Kb block: 47445.82 MB/s SSE & WRITING 4 Kb block: 47792.15 MB/s SSE & WRITING 8 Kb block: 46792.70 MB/s SSE & WRITING 16 Kb block: 47411.45 MB/s SSE & WRITING 32 Kb block: 14151.31 MB/s SSE & WRITING 64 Kb block: 14138.22 MB/s SSE & WRITING 128 Kb block: 14186.34 MB/s SSE & WRITING 256 Kb block: 14076.47 MB/s SSE & WRITING 512 Kb block: 5623.63 MB/s SSE & WRITING 1024 Kb block: 2351.81 MB/s SSE & WRITING 2048 Kb block: 1915.88 MB/s SSE & WRITING 4096 Kb block: 1882.63 MB/s SSE & WRITING 8192 Kb block: 1886.04 MB/s SSE & WRITING 16384 Kb block: 1887.71 MB/s SSE & WRITING 32768 Kb block: 1884.73 MB/s 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: 46786.07 MB/s SSE & READING 2 Kb block: 47461.22 MB/s SSE & READING 4 Kb block: 47794.35 MB/s SSE & READING 8 Kb block: 46800.67 MB/s SSE & READING 16 Kb block: 47395.95 MB/s SSE & READING 32 Kb block: 47701.45 MB/s SSE & READING 64 Kb block: 47772.82 MB/s SSE & READING 128 Kb block: 46840.41 MB/s SSE & READING 256 Kb block: 46883.30 MB/s SSE & READING 512 Kb block: 44452.73 MB/s SSE & READING 1024 Kb block: 44419.80 MB/s SSE & READING 2048 Kb block: 44399.65 MB/s SSE & READING 4096 Kb block: 8824.71 MB/s SSE & READING 8192 Kb block: 6135.76 MB/s SSE & READING 16384 Kb block: 5315.50 MB/s SSE & READING 32768 Kb block: 4973.02 MB/s RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes SSE Copy: 2380.90 MB/s SSE Scale: 2383.04 MB/s SSE Add: 2356.18 MB/s SSE Triad: 2337.27 MB/s --- SSE AVERAGE: 2364.35 MB/s 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: 3020.53 MB/s MMX (nt) & WRITING 2 Kb block: 3020.63 MB/s MMX (nt) & WRITING 4 Kb block: 3011.14 MB/s MMX (nt) & WRITING 8 Kb block: 3012.80 MB/s MMX (nt) & WRITING 16 Kb block: 3015.20 MB/s MMX (nt) & WRITING 32 Kb block: 3008.05 MB/s MMX (nt) & WRITING 64 Kb block: 2992.66 MB/s MMX (nt) & WRITING 128 Kb block: 2960.50 MB/s MMX (nt) & WRITING 256 Kb block: 2896.83 MB/s MMX (nt) & WRITING 512 Kb block: 2774.82 MB/s MMX (nt) & WRITING 1024 Kb block: 2687.01 MB/s MMX (nt) & WRITING 2048 Kb block: 2668.18 MB/s MMX (nt) & WRITING 4096 Kb block: 2666.09 MB/s MMX (nt) & WRITING 8192 Kb block: 2665.59 MB/s MMX (nt) & WRITING 16384 Kb block: 2666.61 MB/s MMX (nt) & WRITING 32768 Kb block: 2660.45 MB/s 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: 23719.80 MB/s MMX (nt) & READING 2 Kb block: 23071.59 MB/s MMX (nt) & READING 4 Kb block: 22730.90 MB/s MMX (nt) & READING 8 Kb block: 22379.88 MB/s MMX (nt) & READING 16 Kb block: 22400.92 MB/s MMX (nt) & READING 32 Kb block: 22374.81 MB/s MMX (nt) & READING 64 Kb block: 22395.44 MB/s MMX (nt) & READING 128 Kb block: 22124.99 MB/s MMX (nt) & READING 256 Kb block: 22165.89 MB/s MMX (nt) & READING 512 Kb block: 21616.44 MB/s MMX (nt) & READING 1024 Kb block: 21608.20 MB/s MMX (nt) & READING 2048 Kb block: 21605.06 MB/s MMX (nt) & READING 4096 Kb block: 9249.11 MB/s MMX (nt) & READING 8192 Kb block: 6942.52 MB/s MMX (nt) & READING 16384 Kb block: 6165.90 MB/s MMX (nt) & READING 32768 Kb block: 5700.68 MB/s 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: 3241.60 MB/s [NTA prefetch] MMX (nt) Scale: 3240.53 MB/s [NTA prefetch] MMX (nt) Add: 3049.50 MB/s [T0 prefetch] MMX (nt) Triad: 3040.13 MB/s [T0 prefetch] --- MMX (nt) AVERAGE: 3142.94 MB/s 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: 3019.64 MB/s SSE & WRITING (nt) 2 Kb block: 3020.49 MB/s SSE & WRITING (nt) 4 Kb block: 3020.04 MB/s SSE & WRITING (nt) 8 Kb block: 3008.34 MB/s SSE & WRITING (nt) 16 Kb block: 3017.86 MB/s SSE & WRITING (nt) 32 Kb block: 3015.20 MB/s SSE & WRITING (nt) 64 Kb block: 3005.93 MB/s SSE & WRITING (nt) 128 Kb block: 2979.09 MB/s SSE & WRITING (nt) 256 Kb block: 2949.70 MB/s SSE & WRITING (nt) 512 Kb block: 2877.08 MB/s SSE & WRITING (nt) 1024 Kb block: 2737.30 MB/s SSE & WRITING (nt) 2048 Kb block: 2661.63 MB/s SSE & WRITING (nt) 4096 Kb block: 2662.28 MB/s SSE & WRITING (nt) 8192 Kb block: 2663.51 MB/s SSE & WRITING (nt) 16384 Kb block: 2657.68 MB/s SSE & WRITING (nt) 32768 Kb block: 2650.38 MB/s RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes SSE & READING (nt) 1 Kb block: 43182.70 MB/s SSE & READING (nt) 2 Kb block: 41083.85 MB/s SSE & READING (nt) 4 Kb block: 40111.63 MB/s SSE & READING (nt) 8 Kb block: 38940.81 MB/s SSE & READING (nt) 16 Kb block: 39022.32 MB/s SSE & READING (nt) 32 Kb block: 39048.49 MB/s SSE & READING (nt) 64 Kb block: 39023.43 MB/s SSE & READING (nt) 128 Kb block: 38369.59 MB/s SSE & READING (nt) 256 Kb block: 38351.82 MB/s SSE & READING (nt) 512 Kb block: 36698.42 MB/s SSE & READING (nt) 1024 Kb block: 36685.14 MB/s SSE & READING (nt) 2048 Kb block: 36660.48 MB/s SSE & READING (nt) 4096 Kb block: 9825.30 MB/s SSE & READING (nt) 8192 Kb block: 7170.05 MB/s SSE & READING (nt) 16384 Kb block: 6269.02 MB/s SSE & READING (nt) 32768 Kb block: 5898.61 MB/s RAMspeed/SMP (Linux) v3.5.0 by Rhett M. Hollander and Paul V. Bolotoff, 2002-09 8Gb per pass mode, 2 processes SSE (nt) Copy: 3349.25 MB/s [NTA prefetch] SSE (nt) Scale: 3291.82 MB/s [NTA prefetch] SSE (nt) Add: 2992.10 MB/s [T0 prefetch] SSE (nt) Triad: 2988.84 MB/s [T0 prefetch] --- SSE (nt) AVERAGE: 3155.50 MB/s
鉴于这服务器的内存是DDR2 800的内存，上面的结果还是可以接受的。
0x04.4 Unix Benchmark
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
Version 5.1.3 Based on the Byte Magazine Unix Benchmark
Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com
1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
1 x Execl Throughput 1 2 3
1 x File Copy 1024 bufsize 2000 maxblocks 1q^H 2 3
1 x File Copy 256 bufsize 500 maxblocks 1 2 3
1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
1 x Process Creation 1 2 3
1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
1 x Shell Scripts (1 concurrent) 1 2 3
1 x Shell Scripts (8 concurrent) 1 2 3
4 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
4 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
4 x Execl Throughput 1 2 3
4 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
4 x File Copy 256 bufsize 500 maxblocks 1 2 3
4 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
4 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
4 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
4 x Process Creation 1 2 3
4 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
4 x Shell Scripts (1 concurrent) 1 2 3
4 x Shell Scripts (8 concurrent) 1 2 3
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: AT166-7318: GNU/Linux
OS: GNU/Linux -- 3.10.0-514.26.1.el7.x86_64 -- #1 SMP Thu Jun 29 16:05:25 UTC 2017
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Atom(TM) CPU D510 @ 1.66GHz (3333.3 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: Intel(R) Atom(TM) CPU D510 @ 1.66GHz (3333.3 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 2: Intel(R) Atom(TM) CPU D510 @ 1.66GHz (3333.3 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 3: Intel(R) Atom(TM) CPU D510 @ 1.66GHz (3333.3 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
07:24:35 up 2:14, 1 user, load average: 0.15, 1.15, 1.34; runlevel 2017-07-01
------------------------------------------------------------------------
Benchmark Run: Sat Jul 01 2017 07:24:35 - 07:52:41
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 6289947.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 889.1 MWIPS (9.9 s, 7 samples)
Execl Throughput 775.9 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 128906.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 36057.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 353356.6 KBps (30.0 s, 2 samples)
Pipe Throughput 367863.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching 23314.1 lps (10.0 s, 7 samples)
Process Creation 1952.9 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1426.1 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 344.8 lpm (60.1 s, 2 samples)
System Call Overhead 557349.1 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 6289947.2 539.0
Double-Precision Whetstone 55.0 889.1 161.7
Execl Throughput 43.0 775.9 180.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 128906.3 325.5
File Copy 256 bufsize 500 maxblocks 1655.0 36057.4 217.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 353356.6 609.2
Pipe Throughput 12440.0 367863.3 295.7
Pipe-based Context Switching 4000.0 23314.1 58.3
Process Creation 126.0 1952.9 155.0
Shell Scripts (1 concurrent) 42.4 1426.1 336.3
Shell Scripts (8 concurrent) 6.0 344.8 574.7
System Call Overhead 15000.0 557349.1 371.6
========
System Benchmarks Index Score 266.8
------------------------------------------------------------------------
Benchmark Run: Sat Jul 01 2017 07:52:41 - 08:20:53
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 15625284.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3070.4 MWIPS (9.4 s, 7 samples)
Execl Throughput 1874.8 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 188573.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 52962.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 531595.9 KBps (30.0 s, 2 samples)
Pipe Throughput 553684.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 110435.3 lps (10.0 s, 7 samples)
Process Creation 4476.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2579.3 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 367.6 lpm (60.3 s, 2 samples)
System Call Overhead 937331.8 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 15625284.5 1338.9
Double-Precision Whetstone 55.0 3070.4 558.3
Execl Throughput 43.0 1874.8 436.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 188573.3 476.2
File Copy 256 bufsize 500 maxblocks 1655.0 52962.8 320.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 531595.9 916.5
Pipe Throughput 12440.0 553684.9 445.1
Pipe-based Context Switching 4000.0 110435.3 276.1
Process Creation 126.0 4476.8 355.3
Shell Scripts (1 concurrent) 42.4 2579.3 608.3
Shell Scripts (8 concurrent) 6.0 367.6 612.7
System Call Overhead 15000.0 937331.8 624.9
========
System Benchmarks Index Score 527.4
从Unix Benchmark的评分来看，这服务器还真和4年前的机子性能差不多。
0x05 结语
虽然性能很差，但对得起这个价格，要知道一年后这服务器就是你的了。但网络这部分确实不适合服务面向国内的用户，不知道是这个数据中心的问题还是所有美国本土的数据中心都很慢。
我测试过很多服务商，面向中国大陆的网络慢可以理解，但是我通过香港的服务器获取krypt上的文件，也一样很慢，这就不太正常了。
0x06 捐赠
因为资金的关系，新加坡机房的测试计划在7月下旬进行。
如果你觉得我的文章或视频对你有帮助，欢迎对我进行捐赠，捐赠的资金我会用来购买测试用的服务器进行测试。
0x07 测试
这个测试服务器有效期到2017年7月30日，在此之前可以通过以下地址进行下载速度测试：
ip1.us.t1.bever.in ip2.us.t1.bever.in ip3.us.t1.bever.in ip4.us.t1.bever.in ip5.us.t1.bever.in
可以通过以下地址进行PHP探针测试：
ip1.us.t1.bever.in/php/tz.php ip2.us.t1.bever.in/php/tz.php ip3.us.t1.bever.in/php/tz.php ip4.us.t1.bever.in/php/tz.php ip5.us.t1.bever.in/php/tz.php
通过以下地址查看phpinfo：
ip1.us.t1.bever.in/php/phpinfo.php ip2.us.t1.bever.in/php/phpinfo.php ip3.us.t1.bever.in/php/phpinfo.php ip4.us.t1.bever.in/php/phpinfo.php ip5.us.t1.bever.in/php/phpinfo.php
