华为云两款2X4G鸡比较

我曾在今年7月多上了华为云的双核4G1M学生机,当时是s2.large.2而且只有广州数据中心,今天,我再次进入华为云的学生优惠页,发现我居然还能买2x4g学生机,活动内容基本一样,但这次不同的是,型号变成了t6.large.2,而且有上海、北京可选(现在还有云数据库可供选购),因为我腾讯和阿里的机子都在上海,所以这次我上了北京的年付车

首先,说一下s2.large.2和t6.large.2的区别

t6.large.2
s2.large.2

可以看出,t6有性能限制,超过基准就要算CPU积分,积分用完就会强制限制到基准(积分啥的我完全搞不懂)

直接放跑分结果把,虽然不是在一个地区,但应该还是有参考性的

这是s2的,性能还过得去,就是不知道为什么speedtest跑不动
------------------------------------------------------------------------
Benchmark Run: Tue Oct 29 2019 19:52:14 - 12:20:51
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       33486251.2 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3802.8 MWIPS (13.2 s, 7 samples)
Execl Throughput                               4723.8 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        859295.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          226276.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2355875.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1570786.9 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  89506.2 lps   (10.0 s, 7 samples)
Process Creation                              10460.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   7195.0 lpm   (60.3 s, 2 samples)
Shell Scripts (8 concurrent)                   1553.7 lpm   (60.0 s, 2 samples)
System Call Overhead                        2015708.8 lps   (10.0 s, 7 samples)
 
System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   33486251.2   2869.4
Double-Precision Whetstone                       55.0       3802.8    691.4
Execl Throughput                                 43.0       4723.8   1098.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     859295.0   2169.9
File Copy 256 bufsize 500 maxblocks            1655.0     226276.3   1367.2
File Copy 4096 bufsize 8000 maxblocks          5800.0    2355875.5   4061.9
Pipe Throughput                               12440.0    1570786.9   1262.7
Pipe-based Context Switching                   4000.0      89506.2    223.8
Process Creation                                126.0      10460.1    830.2
Shell Scripts (1 concurrent)                     42.4       7195.0   1696.9
Shell Scripts (8 concurrent)                      6.0       1553.7   2589.6
System Call Overhead                          15000.0    2015708.8   1343.8
                                                                   ========
System Benchmarks Index Score                                        1349.2
 
------------------------------------------------------------------------
Benchmark Run: Tue Oct 29 2019 12:20:51 - 12:49:32
2 CPUs in system; running 2 parallel copies of tests
 
Dhrystone 2 using register variables       65907398.7 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     7605.8 MWIPS (13.2 s, 7 samples)
Execl Throughput                               9185.6 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1147725.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          303610.9 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       3320440.7 KBps  (30.0 s, 2 samples)
Pipe Throughput                             3039240.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 499038.8 lps   (10.0 s, 7 samples)
Process Creation                              22374.7 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  11306.6 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1647.2 lpm   (60.0 s, 2 samples)
System Call Overhead                        3151873.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   65907398.7   5647.6
Double-Precision Whetstone                       55.0       7605.8   1382.9
Execl Throughput                                 43.0       9185.6   2136.2
File Copy 1024 bufsize 2000 maxblocks          3960.0    1147725.8   2898.3
File Copy 256 bufsize 500 maxblocks            1655.0     303610.9   1834.5
File Copy 4096 bufsize 8000 maxblocks          5800.0    3320440.7   5724.9
Pipe Throughput                               12440.0    3039240.7   2443.1
Pipe-based Context Switching                   4000.0     499038.8   1247.6
Process Creation                                126.0      22374.7   1775.8
Shell Scripts (1 concurrent)                     42.4      11306.6   2666.6
Shell Scripts (8 concurrent)                      6.0       1647.2   2745.3
System Call Overhead                          15000.0    3151873.1   2101.2
                                                                   ========
System Benchmarks Index Score                                        2429.5

Superbench
----------------------------------------------------------------------
 Superbench.sh -- https://www.oldking.net/350.html
 Mode  : Standard    Version : 1.1.5
 Usage : wget -qO- git.io/superbench.sh | bash
----------------------------------------------------------------------
 CPU Model            : Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz
 CPU Cores            : 2 Cores @ 2394.454 MHz x86_64
 CPU Cache            : 35840 KB 
 OS                   : CentOS 7.6.1810 (64 Bit) KVM
 Kernel               : 5.3.7-1.el7.elrepo.x86_64
 Total Space          : 3.2 GB / 40.0 GB 
 Total RAM            : 233 MB / 3938 MB (745 MB Buff)
 Total SWAP           : 0 MB / 0 MB
 Uptime               : 0 days 1 hour 55 min
 Load Average         : 0.00, 0.00, 0.16
 TCP CC               : bbr
 ASN & ISP            : AS55990, Huawei Cloud Service
 Organization         : Huawei Cloud Service
 Location             : Guangzhou, China / CN
 Region               : Guangdong
----------------------------------------------------------------------
 I/O Speed( 1.0GB )   : 166 MB/s
 I/O Speed( 1.0GB )   : 157 MB/s
 I/O Speed( 1.0GB )   : 179 MB/s
 Average I/O Speed    : 167.3 MB/s
----------------------------------------------------------------------
 Node Name        Upload Speed      Download Speed      Latency     
 Speedtest.net    2.28 Mbit/s       84.12 Mbit/s        43.286 ms  
 Wuhan     CT     1.95 Mbit/s       71.59 Mbit/s        -          
----------------------------------------------------------------------
 Finished in  : 6 min 1 sec
 Timestamp    : 2019-10-29 13:42:05 GMT+8
 Results      : /root/superbench.log
----------------------------------------------------------------------
这是t6的,性能要差点,但也过得去,ip貌似是广播上海的
------------------------------------------------------------------------
Benchmark Run: Tue Oct 29 2019 12:21:24 - 12:49:30
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       33006253.8 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3927.5 MWIPS (9.8 s, 7 samples)
Execl Throughput                               4615.3 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        670588.1 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          183666.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1806455.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1057433.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  75775.1 lps   (10.0 s, 7 samples)
Process Creation                               9376.1 lps   (30.4 s, 2 samples)
Shell Scripts (1 concurrent)                   7090.6 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1458.5 lpm   (60.0 s, 2 samples)
System Call Overhead                         928061.5 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   33006253.8   2828.3
Double-Precision Whetstone                       55.0       3927.5    714.1
Execl Throughput                                 43.0       4615.3   1073.3
File Copy 1024 bufsize 2000 maxblocks          3960.0     670588.1   1693.4
File Copy 256 bufsize 500 maxblocks            1655.0     183666.3   1109.8
File Copy 4096 bufsize 8000 maxblocks          5800.0    1806455.5   3114.6
Pipe Throughput                               12440.0    1057433.7    850.0
Pipe-based Context Switching                   4000.0      75775.1    189.4
Process Creation                                126.0       9376.1    744.1
Shell Scripts (1 concurrent)                     42.4       7090.6   1672.3
Shell Scripts (8 concurrent)                      6.0       1458.5   2430.8
System Call Overhead                          15000.0     928061.5    618.7
                                                                   ========
System Benchmarks Index Score                                        1118.3

------------------------------------------------------------------------
Benchmark Run: Tue Oct 29 2019 12:49:30 - 13:17:34
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       66023784.7 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     7831.6 MWIPS (9.8 s, 7 samples)
Execl Throughput                               8116.9 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1067156.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          279506.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2916760.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             2090724.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 419397.7 lps   (10.0 s, 7 samples)
Process Creation                              14908.3 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  10863.4 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1540.2 lpm   (60.0 s, 2 samples)
System Call Overhead                        1620902.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   66023784.7   5657.6
Double-Precision Whetstone                       55.0       7831.6   1423.9
Execl Throughput                                 43.0       8116.9   1887.6
File Copy 1024 bufsize 2000 maxblocks          3960.0    1067156.4   2694.8
File Copy 256 bufsize 500 maxblocks            1655.0     279506.3   1688.9
File Copy 4096 bufsize 8000 maxblocks          5800.0    2916760.0   5028.9
Pipe Throughput                               12440.0    2090724.8   1680.6
Pipe-based Context Switching                   4000.0     419397.7   1048.5
Process Creation                                126.0      14908.3   1183.2
Shell Scripts (1 concurrent)                     42.4      10863.4   2562.1
Shell Scripts (8 concurrent)                      6.0       1540.2   2566.9
System Call Overhead                          15000.0    1620902.1   1080.6
                                                                   ========
System Benchmarks Index Score                                        2038.8

Superbench
----------------------------------------------------------------------
 Superbench.sh -- https://www.oldking.net/350.html
 Mode  : Standard    Version : 1.1.5
 Usage : wget -qO- git.io/superbench.sh | bash
----------------------------------------------------------------------
 CPU Model            : Intel(R) Xeon(R) Gold 6161 CPU @ 2.20GHz
 CPU Cores            : 2 Cores @ 2200.000 MHz x86_64
 CPU Cache            : 30976 KB 
 OS                   : CentOS 7.6.1810 (64 Bit) KVM
 Kernel               : 5.3.7-1.el7.elrepo.x86_64
 Total Space          : 3.2 GB / 40.0 GB 
 Total RAM            : 239 MB / 3938 MB (674 MB Buff)
 Total SWAP           : 0 MB / 0 MB
 Uptime               : 0 days 0 hour 57 min
 Load Average         : 9.54, 7.40, 4.25
 TCP CC               : bbr
 ASN & ISP            : AS55990, Huawei Cloud Service
 Organization         : Huawei Cloud Service
 Location             : Shanghai, China / CN
 Region               : Shanghai
----------------------------------------------------------------------
 I/O Speed( 1.0GB )   : 182 MB/s
 I/O Speed( 1.0GB )   : 156 MB/s
 I/O Speed( 1.0GB )   : 156 MB/s
 Average I/O Speed    : 164.7 MB/s
----------------------------------------------------------------------
 Node Name        Upload Speed      Download Speed      Latency     
 Speedtest.net    1.80 Mbit/s       84.02 Mbit/s        47.467 ms  
 Fast.com         0.00 Mbit/s       0.8 Mbit/s          -           
 Wuhan     CT     1.85 Mbit/s       84.62 Mbit/s        -          
 Hangzhou  CT     1.56 Mbit/s       78.70 Mbit/s        -          
 Lanzhou   CT     1.75 Mbit/s       48.04 Mbit/s        -          
 Shanghai  CU     1.58 Mbit/s       93.58 Mbit/s        -          
 Heifei    CU     1.94 Mbit/s       64.39 Mbit/s        -          
 Xizang    CM     1.55 Mbit/s       70.12 Mbit/s        -          
----------------------------------------------------------------------
 Finished in  : 6 min 25 sec
 Timestamp    : 2019-10-29 13:24:13 GMT+8
 Results      : /root/superbench.log
----------------------------------------------------------------------

在测试过程中,后台显示s2最高跑到了94%的CPU占用率,而t6最高只有80%,积分也用了不少…

说不清楚是变坑了还是怎么样…

告别Vultr…

Vultr是我接触比较早的服务商(大概就在17年吧,那时日本2.5刀还有货),因为较为合理的定价(日本5刀还行),有保障的性能(unixbench常年1300),便捷的快照,稳定的服务,所以我一直用到了现在

但如今各大亚太主机商定价大幅下跌(据我所拥有的主机定价来说,1G内存月付最低可至日本2刀,香港5刀,甚至有1点多刀的香港CN2),vu对我而言再也香不起来了

Vultr日本 11:10实测(5刀/月)
AWS Lightsail日本11:10实测(3块/月)
某主机商IIJ日本11:10实测(2刀/月)
某主机商CMI香港11:10实测(5刀/月)
某主机商CN2香港11:10实测(5刀/月)

vu日本网络常年爆炸,加之我的站点也多数迁至虚拟主机,从一开始的建站主力,到现在日常吃灰,在我眼里,vu早已失去了曾经的价值,虽然前后加起来我在vu也花去了100多刀,出于情怀,我还是一直续费到了今天,但如今,秉持节流原则,我决定不再续费vu日本,从前几日开始,我就在进行文件转移,oneindex,小鸡探针,vu上最后留下的网站也悉数转移

但当我真按下Destroy,看着空空的Products页,还是有些不舍

就这样吧…

日常

入手3台学生机…来个对比…

阿里云腾讯云华为云
配置100%1X 2G1X 2G1X 2G / 2X 4G
网络1M/1000G 5M1M1M
储存40G50G40G
地域上海/北京上海/北京广州
价格9.5/月10/月9/19/月
续费24岁前无限认证后续2次,之后可抢购续费资格可续3次
备注取消后可重订取消即放弃资格取消即放弃资格

华为云配置不错,阿里可长期持有,腾讯像是促销(雾),销售策略像逼着你一直年付…

目前这3台处于空转状态,但实测的话阿里云的性能是强于腾讯的,它的下行带宽也未限速(其他两家都限100M)就是上行让我非常难办…

购入2个A1全局管理员,我现在可以开20000个A1哟…

对了,哪吒吹爆,看就是了

咕咕咕

本来我最近都不想写的,结果一封邮件把我炸了出来

哇,VPSFast复活啦…md当时买就是我鬼迷心窍…

付前:这香港真便宜,8刀年付付10个月,还打六折,比特币支付还有额外20%OFF,不买不是人,买买买

付完后:有点小后悔,是不是灵车,可车门焊死了,下不了(比特币支付通常无法退款…)

许久之后:这香港怎么还tm没开通,不行,我要工单伺候

我:香港怎么还没好

许久之后,他们:快了快了

又是许久,我:我tm好急

他们:啊啊,香港没货了,你看看别的区可以么

我:有什么区

许久,他们:毛熊呀,棒子呀

我:我选择棒子…

然后还不错,当天就开了

我:哈哈哈,我也是有韩国VPS的人了…

然而现实给了我沉重一击…没多久它就…就Down了

我等呀等,还在侥幸这可能只是意外,最后忍无可忍又工单伺候(现在这工单系统里已经没了,只有邮件记录)

好,我等(一个月的时候Down过一次)

然后实在忍无可忍,语气加重(现在也没了,其实,估计是系统还原了)

结果石沉大海,再开网站就各种PHP错误

估计它跑路了,然而没多久,又复活了…

哟,当你的合伙人?(XSWL)还想骗我第二次?

趁着网站正常,我赶紧再来工单一封…

我:你们服务器怎么了,死了3月了

他们:我们遇到点小问题

我:啥时好,我可不可以换地区

他们:快了快了…

然后等呀等…就在我正准备再次工单时,咦,官网怎么打不开了?

吓得我赶紧论坛查一波,然后就得知跑路已久的消息…

你妈哟,凉了凉了…

本来我都忘了这事,结果,你tm还有脸催我续费?

千万别上这家的车…

就酱,我要冷静…

无题

啊啊啊,报志愿好烦…

好多事,但我现在在肝夏促,夏促完再说…

想买十年大会员,-332还行啦…(好贵)

光猫是OpenWRT

刚发现DignusData DOWN了然后我这CDN就崩溃了…(已修复)

发现光猫是OpenWRT,但是,它阉了wget,opkg……没wget我怎么装opkg….

另,LeaseWeb CDN已加入,但貌似没有亚太节点…

有v6了

今天突然连不上网,就向移动报了个修,然后他们就换了光猫,换好之后我在后台发现有v6了…但是,桥接到小米路由后,经过我多次尝试,分到的/64就算分给了下级路由(发现是k2p无法再分…),也无法使用…(下次再说,反正公网ip也挂在上面)所以,我的公网变成v4/v6双栈了…

然后,我现在要进行资源清点,把不需要的鸡都删了….(咕咕咕,今天没干…)

就酱吧(对了,LeaseWeb CDN正在测试中,稳定的话加入CDN豪华午餐…)

咕咕咕

考完好几天了……

无聊了好几天了…所以博客名改了…咕咕咕,不想写了…

不过八方旅人是真好玩,发张壁纸(未压缩版自行删去参数…)

买了一加7 pro,体验良好

不知不觉就快一年了,我的.me域名就要到期了…留着主要是Gandi的DNS支持多ip,而dnspod的要钱…所以CDN用的.me…

对了戈登云的NAT升级了独立ip…(性价比高了一大坨)

然而,戈登云太jb坑了,就ping了一下,说我被DDOS,封机24h(感觉以后有的逼要撕…)已妥善解决

然后AlphaRack跑路还不忘收割一波aff

就酱吧…我都闲出病了…