源站网干货教程 2020-08-18

【行业禁忌？】小鸡的CPU跑分多少算合格？

不同商家对于小鸡CPU的限制差别挺大的，就算是我们自己对于CPU的限制也是根据产品的定位而定的。

也没啥好避讳的，我们的EPYC的限制是98%，KVM是95%，VDS是80%，RAM是50%，NAT小鸡是30%
后面我会逐步把各个限制下的sysbench跑分结果和宿主机的情况发出来，供大家参考

大家也可以测测看自己小鸡的跑分结果，发在这里，行业透明才有利于行业的发展
一次跑分说明不了什么，可以有事没事跑个分，样本量大了就有代表性了（回复都整理到第二页了，可以看看）

CentOS：

yum install epel-release -y
yum install sysbench -y
sysbench cpu run

复制代码

Ubuntu：

apt-get install sysbench
sysbench cpu run

复制代码

多线程的小鸡请加上线程参数进行测试（替换N为自己的核数）
–num-threads=N

另外也可以参考unixbench（比较耗时，测试更加偏向全性能）：

wget –no-check-certificate https://github.com/zq/unixbench/raw/master/unixbench.sh
chmod +x unixbench.sh
./unixbench.sh

复制代码

——-第二页为各位回复汇总——–
——-其后为我们的实际测评——–

网友回复：

CloudRaft：等会回来贴分数

van：先关注下

嗷嗷：推荐睿数据的服务器，G口机器不超开，ruidata.net。

laogui：我推荐UnixBench跑分，Vultr原先也是标的UnixBench跑分，现在改Geekbench的跑分了。 http://www.unixbench.org/ 1. 辣鸡PR，单核1G：最开始300多分，后面跑分500出头。 2. 搬瓦工双11款，双核2G，单核881/双核1337 3. 给同事推荐的Offer区VPS，半翻车了，跑分单核285/双核536

403_Forbidden： VDS怎么比KVM低

micto：一时的跑分说明不了什么

nico：挺好的啊铁汁

gzchenjz：路过绑定

注册： 90qh [root[url=home.php?mod=space&uid=22998]@localhost[/url] ~]# sysbench cpu run sysbench 1.0.17 (using system LuaJIT 2.0.4) Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 258.71 General statistics: total time: 10.0010s total number of events: 2589 Latency (ms): min: 2.87 avg: 3.86 max: 11.81 95th percentile: 4.82 sum: 9992.41 Threads fairness: events (avg/stddev): 2589.0000/0.00 execution time (avg/stddev): 9.9924/0.00 复制代码瓦工 [root[url=home.php?mod=space&uid=16978]@host[/url] ~]# sysbench cpu run sysbench 1.0.17 (using system LuaJIT 2.0.4) Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 781.16 General statistics: total time: 10.0010s total number of events: 7814 Latency (ms): min: 1.21 avg: 1.28 max: 4.90 95th percentile: 1.34 sum: 9989.26 Threads fairness: events (avg/stddev): 7814.0000/0.00 execution time (avg/stddev): 9.9893/0.00 复制代码

注册：云筏4核KVM # sysbench cpu run sysbench 1.0.17 (using system LuaJIT 2.0.4) Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 964.56 General statistics: total time: 10.0009s total number of events: 9648 Latency (ms): min: 0.93 avg: 1.04 max: 6.35 95th percentile: 1.08 sum: 9993.41 Threads fairness: events (avg/stddev): 9648.0000/0.00 execution time (avg/stddev): 9.9934/0.00 复制代码

king51：云筏1核NAT小鸡 # sysbench cpu run sysbench 1.0.17 (using system LuaJIT 2.0.4) Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 138.45 General statistics: total time: 10.0106s total number of events: 1387 Latency (ms): min: 0.93 avg: 7.20 max: 38.05 95th percentile: 10.27 sum: 9991.73 Threads fairness: events (avg/stddev): 1387.0000/0.00 execution time (avg/stddev): 9.9917/0.00 复制代码

enulann：套路云轻量 # sysbench cpu run sysbench 1.0.17 (using system LuaJIT 2.0.4) Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 858.41 General statistics: total time: 10.0010s total number of events: 8587 Latency (ms): min: 1.08 avg: 1.16 max: 5.43 95th percentile: 1.34 sum: 9985.90 Threads fairness: events (avg/stddev): 8587.0000/0.00 execution time (avg/stddev): 9.9859/0.00 复制代码

enulann：绿云大阪 CPU speed: events per second: 861.03 General statistics: total time: 10.0017s total number of events: 8614 Latency (ms): min: 1.09 avg: 1.16 max: 28.63 95th percentile: 1.23 sum: 9990.83 Threads fairness: events (avg/stddev): 8614.0000/0.00 execution time (avg/stddev): 9.9908/0.00 复制代码 ion2019 Threads started! CPU speed: events per second: 789.12 General statistics: total time: 10.0011s total number of events: 7894 Latency (ms): min: 1.16 avg: 1.27 max: 22.37 95th percentile: 1.39 sum: 9993.16 Threads fairness: events (avg/stddev): 7894.0000/0.00 execution time (avg/stddev): 9.9932/0.00 复制代码 CC15刀 CPU speed: events per second: 428.16 General statistics: total time: 10.0017s total number of events: 4284 Latency (ms): min: 1.55 avg: 2.33 max: 62.88 95th percentile: 2.48 sum: 9977.21 Threads fairness: events (avg/stddev): 4284.0000/0.00 execution time (avg/stddev): 9.9772/0.00 复制代码腾讯云北京 Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 942.58 General statistics: total time: 10.0005s total number of events: 9428 Latency (ms): min: 1.01 avg: 1.06 max: 2.17 95th percentile: 1.08 sum: 9986.02 Threads fairness: events (avg/stddev): 9428.0000/0.00 execution time (avg/stddev): 9.9860/0.00 复制代码腾讯云香港轻量 Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 906.46 General statistics: total time: 10.0042s total number of events: 9070 Latency (ms): min: 1.05 avg: 1.10 max: 16.86 95th percentile: 1.18 sum: 9991.56复制代码 PR7.25 Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 587.91 General statistics: total time: 10.0043s total number of events: 5883 Latency (ms): min: 1.36 avg: 1.69 max: 37.56 95th percentile: 2.14 sum: 9945.14 复制代码

enulann：这个命令仅仅是单线程。

Zxwax：云筏2核kvm hetzner sysbench 1.0.17 (using system LuaJIT 2.0.4) Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 954.91 General statistics: total time: 10.0005s total number of events: 9551 Latency (ms): min: 0.93 avg: 1.05 max: 8.21 95th percentile: 1.06 sum: 9991.46 Threads fairness: events (avg/stddev): 9551.0000/0.00 execution time (avg/stddev): 9.9915/0.00 复制代码这是某mjj开的8核nat sysbench 1.0.17 (using system LuaJIT 2.0.4) Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 518.05 General statistics: total time: 10.0013s total number of events: 5184 Latency (ms): min: 1.78 avg: 1.93 max: 5.86 95th percentile: 2.48 sum: 9989.99 Threads fairness: events (avg/stddev): 5184.0000/0.00 execution time (avg/stddev): 9.9900/0.00 复制代码顺带跑了个ikoula家的r9 3900x独服 sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2) Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 2278.33 General statistics: total time: 10.0004s total number of events: 22786 Latency (ms): min: 0.43 avg: 0.44 max: 0.58 95th percentile: 0.45 sum: 9998.19 Threads fairness: events (avg/stddev): 22786.0000/0.00 execution time (avg/stddev): 9.9982/0.00 复制代码

012： CPU speed: events per second: 1480.20 General statistics: total time: 10.0005s total number of events: 14804 Latency (ms): min: 0.52 avg: 0.67 max: 8.40 95th percentile: 1.55 sum: 9981.84 Threads fairness: events (avg/stddev): 14804.0000/0.00 execution time (avg/stddev): 9.9818/0.00

yimi233： geekbench 5我觉得最起码得跑到500分以上。

micms：虽然没买过你家鸡但一直觉得楼主的言语很坦诚

yizhuohao7： virmach家的： Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 556.06 General statistics: total time: 10.0017s total number of events: 5563 Latency (ms): min: 1.55 avg: 1.79 max: 31.70 95th percentile: 1.86 sum: 9952.60 Threads fairness: events (avg/stddev): 5563.0000/0.00 execution time (avg/stddev): 9.9526/0.00

2543303982： Hello, Kotlin

ApkB： Geekbench 单核875/双核1736

js攻城狮： Benchmark Run: Mon Aug 17 2020 21:50:47 – 22:19:11 1 CPU in system; running 1 parallel copy of tests Dhrystone 2 using register variables 26227414.0 lps (10.0 s, 7 samples) Double-Precision Whetstone 3777.1 MWIPS (9.8 s, 7 samples) Execl Throughput 2418.2 lps (30.0 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 194659.8 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 49822.0 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 577919.4 KBps (30.0 s, 2 samples) Pipe Throughput 292582.2 lps (10.0 s, 7 samples) Pipe-based Context Switching 64489.9 lps (10.0 s, 7 samples) Process Creation 5467.4 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 4733.7 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 596.8 lpm (60.1 s, 2 samples) System Call Overhead 258623.7 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 26227414.0 2247.4 Double-Precision Whetstone 55.0 3777.1 686.8 Execl Throughput 43.0 2418.2 562.4 File Copy 1024 bufsize 2000 maxblocks 3960.0 194659.8 491.6 File Copy 256 bufsize 500 maxblocks 1655.0 49822.0 301.0 File Copy 4096 bufsize 8000 maxblocks 5800.0 577919.4 996.4 Pipe Throughput 12440.0 292582.2 235.2 Pipe-based Context Switching 4000.0 64489.9 161.2 Process Creation 126.0 5467.4 433.9 Shell Scripts (1 concurrent) 42.4 4733.7 1116.4 Shell Scripts (8 concurrent) 6.0 596.8 994.7 System Call Overhead 15000.0 258623.7 172.4 ======== System Benchmarks Index Score 521.3

hcyme：辣鸡PR老套餐根本没有CPU限制，直接给100%，就看谁抢得过谁…… 新套餐统一50%，这个后续会往上调，我正在研究动态降频方法。比如每隔10分钟检测一次，最早给跑100%，如果10分钟跑满降频10%，以此类推。

yem：老套餐换新套餐，降带宽、降流量的事情，这可是马蜂窝。

开腥小站长：老套餐回收没多大利润，还特别麻烦，所以商量了下直接集体升Virtualizor。升完以后先给几个高负载节点限制CPU，把负载降下来再说。后续PR大概会划分产品线，PR-V是往国际上做，走高端。之前的那些修好负载问题，可以出一些无节操促销。经过测试一台母鸡放280台小鸡没啥问题。

micto： sysbench cpu run Unknown command: cpu. Usage: sysbench [general-options]… –test=

[test-options]… command 为什么我命令行每次都失败只能这样又看不到CPU分数 root@cont59:~# sysbench –num-threads=4 –test=cpu run sysbench 0.4.12: multi-threaded system evaluation benchmark Running the test with following options: Number of threads: 4 Doing CPU performance benchmark Threads started! Done. Maximum prime number checked in CPU test: 10000 Test execution summary: total time: 1.6933s total number of events: 10000 total time taken by event execution: 6.7672 per-request statistics: min: 0.59ms avg: 0.68ms max: 16.35ms approx. 95 percentile: 0.87ms Threads fairness: events (avg/stddev): 2500.0000/221.02 execution time (avg/stddev): 1.6918/0.00 root@cont59:~#

开腥小站长：那还是动态降频比较好，这个技术貌似没几家有，我得研究研究。（不对啊我的职务是销售吧……

micto：搬瓦工双11 WARNING: –num-threads is deprecated, use –threads instead sysbench 1.0.17 (using system LuaJIT 2.1.0-beta3) Running the test with following options: Number of threads: 2 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 1134.74 General statistics: total time: 10.0018s total number of events: 11353 Latency (ms): min: 1.54 avg: 1.76 max: 23.80 95th percentile: 2.03 sum: 19976.20 Threads fairness: events (avg/stddev): 5676.5000/78.50 execution time (avg/stddev): 9.9881/0.00 复制代码

开腥小站长： pr第一批鸡鸡（三号母鸡） sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3) Running the test with following options: Number of threads: 2 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 1413.85 General statistics: total time: 10.0012s total number of events: 14144 Latency (ms): min: 1.18 avg: 1.41 max: 12.25 95th percentile: 1.52 sum: 19960.98 Threads fairness: events (avg/stddev): 7072.0000/2.00 execution time (avg/stddev): 9.9805/0.00 复制代码

面包狗： ~# sysbench cpu run sysbench 1.0.19 (using bundled LuaJIT 2.1.0-beta2) Running the test with following options: Number of threads: 1 Initializing random number generator from current time Prime numbers limit: 10000 Initializing worker threads… Threads started! CPU speed: events per second: 1105.90 General statistics: total time: 10.0001s total number of events: 11061 Latency (ms): min: 0.75 avg: 0.90 max: 3.96 95th percentile: 1.01 sum: 9992.11 Threads fairness: events (avg/stddev): 11061.0000/0.00 execution time (avg/stddev): 9.9921/0.00

面包狗： ion2019CPU speed: events per second: 788.60 General statistics: total time: 10.0006s total number of events: 7888 Latency (ms): min: 1.17 avg: 1.27 max: 24.28 95th percentile: 1.30 sum: 9994.36 Threads fairness: events (avg/stddev): 7888.0000/0.00 execution time (avg/stddev): 9.9944/0.00 复制代码 GGC hkCPU speed: events per second: 560.54 General statistics: total time: 10.0017s total number of events: 5608 Latency (ms): min: 1.41 avg: 1.78 max: 13.17 95th percentile: 2.57 sum: 9986.80 复制代码水墨云hk建站Threads started! CPU speed: events per second: 699.98 General statistics: total time: 10.0011s total number of events: 7003 Latency (ms): min: 1.03 avg: 1.43 max: 9.63 95th percentile: 1.96 sum: 9994.64 Threads fairness: events (avg/stddev): 7003.0000/0.00 execution time (avg/stddev): 9.9946/0.00 复制代码套路云上海CPU speed: events per second: 867.17 General statistics: total time: 10.0007s total number of events: 8674 Latency (ms): min: 1.07 avg: 1.15 max: 71.47 95th percentile: 1.25 sum: 9990.68 复制代码

注意：本站只对HOSTLOC帖子进行收集和归纳，并不审核其内容，若其中内容侵犯到您的权益，请联系长站删除！！本文由IP索引采集整理发布！！ipipn.com欢迎友情链接！！！

相关说明：

1、VIP会员无限制任意下载，免积分。立即前往开通>>

2、下载积分可通过日常签到、绑定邮箱以及积分兑换等途径获得！

3、本站资源大多存储在云盘，如出现链接失效请评论反馈,如有密码,均为:www.ipipn.com。

4、所有站内资源仅供学习交流使用。未经原版权作者许可，禁止用于任何商业环境，否则后果自负。为尊重作者版权，请购买正版作品。