老刘博客
本文目录
显示
UCloud云服务器在2020年加大了推广力度，实打实地给出了不少让利，对于有云服务器需求的用户来说，这是好事！那么UCloud云服务器性能、速度到底怎么样呢？老刘博客之前文章当中对UCloud香港云服务器的性能、速度做了比较详细的测评！具体可参见《UCloud香港快杰云主机速度及性能测评附如何领券购买教程》，《到底什么样需求的用户需要UCloud Scalable processors 快杰云主机？》等。从测试结果来说，UCloud属于云服务上市企业，云服务器品质自然没有大的问题，符合高性能企业级用途标准，带宽质量、服务器稳定性都还很不错的。老刘博客也是在有体验的，目前老刘博客使用的也是UCloud香港云服务器产品！打开速度、在线率及晚高峰速度大家只要关注下博客就知道了！
目前UCloud云服务器分为快杰型云主机和通用型云主机等（普通用户需求这两种就够了），具体分类可参见老刘博客前文《UCloud优刻得云主机机型与CPU平台》。快杰云主机是近期UCloud主推的云服务器产品，他们家国内云服务器、海外云服务器特价促销方案基本都已经换成了快杰云主机方案！综合上面博主测评香港快杰云主机和香港通用云主机对比结果来看，同CPU类型的两款云服务器在性能方面快杰云主机要稍微好一点点，大的什么突出表现是没有的，但是快杰型云主机官网原价比通用型云主机更便宜！具体价格对比老刘博客在《UCloud香港快杰型云主机与通用型云主机性能测评对比》有过介绍。这里我们重点再来看看UCloud国内快杰云服务器速度及性能怎么样？
一、UCloud国内快杰云主机促销策略
UCloud国内快杰云主机促销方案主打的是Intel CPU产品，活动不限新老用户，但是限新购，每个用户限购买一台！有国内机房：北京、上海、广州三个地区可以选择！最新的促销方案如下，相比之前进行了调整，1核2G内存2M带宽方案放宽了购买期限，也可以一次性购入三年了，三年仅需258元！平均仅需7元/月起！并且是BGP带宽云服务器，国内目前云计算主流商家当中应该是最便宜的了！
购买UCloud云主机过程当中一定要记得看准配置及机房，UCloud国内快杰云主机促销的地域有北京二E区、上海二A区、广州B区，另外就是配置、带宽、硬盘都要选对。更多细节参见《UCloud北上广机房1C2G2M快杰云主机首年88元3年258元》。
| CPU | 内存 | 硬盘 | BGP带宽/流量 | 价格 | 链接 |
| 1核 | 2G | 40G+20G | 2Mbps/无限 | 88元/首年 258元/3年 | 点击直达 |
| 2核 | 4G | 40G+100G | 5Mbps/无限 | 468元/首年 1398元/3年 | 点击直达 |
| 4核 | 8G | 40G+100G | 5Mbps/无限 | 888元/首年 2658元/3年 | 点击直达 |
| 8核 | 16G | 40G+100G | 5Mbps/无限 | 3888元/首年 11658元/3年 | 点击直达 |
| 8核 | 16G | 40G+100G | 10Mbps/无限 | 7888元/首年 23658元/3年 | 点击直达 |
二、国内快杰云主机和通用云主机综合测评对比
老刘博客自己开通了一台2核4G内存5M带宽配置的上海A区快杰云主机和一款同配置同区域的通用云主机，选择的都是Intel CPU，建议有需求的话也选择Intel CPU！下面我们一起来看看UCloud国内快杰云服务器速度、性能怎么样，测试结果仅供参考。
1、国内电信、联通、移动ping测试结果
BGP带宽，136个节点平均延迟27.2 毫秒，低延迟没毛病，海外服务器是无法达到这个水平的（比如老刘博客的延迟大概是4-50毫秒）。
| 线路 | 节点数目 | 最快节点 | 最慢节点 | 平均响应 |
|---|---|---|---|---|
| 全部 | 136 | 上海地面通(BGP) 4.0 毫秒 | 新疆乌鲁木齐(天翼云) 77.7 毫秒 | 27.2 毫秒 |
| 联通线路 | 16 | 江苏镇江(联通) 9.0 毫秒 | 云南昆明(联通) 47.4 毫秒 | 23.9 毫秒 |
| 电信线路 | 59 | 上海(天翼云四区) 4.7 毫秒 | 新疆乌鲁木齐(天翼云) 77.7 毫秒 | 29.3 毫秒 |
| 移动线路 | 10 | 江苏常州(移动) 7.0 毫秒 | 广东茂名(移动) 45.7 毫秒 | 30.0 毫秒 |
| 东北地区 | 11 | 辽宁大连(电信) 39.5 毫秒 | 吉林长春(铁通) 65.9 毫秒 | 47.9 毫秒 |
| 华北地区 | 18 | 天津(腾讯云) 20.5 毫秒 | 北京(阿里云) 34.7 毫秒 | 27.8 毫秒 |
| 西北地区 | 6 | 陕西西安(天翼云二区) 26.9 毫秒 | 新疆乌鲁木齐(天翼云) 77.7 毫秒 | 43.7 毫秒 |
| 西南地区 | 15 | 重庆(腾讯云) 31.9 毫秒 | 四川成都(AS9929) 48.4 毫秒 | 38.9 毫秒 |
| 华中地区 | 11 | 湖北武汉(天翼云) 18.7 毫秒 | 河南新乡(7yc.com) 42.9 毫秒 | 27.4 毫秒 |
| 华东地区 | 49 | 上海地面通(BGP) 4.0 毫秒 | 江苏宿迁(移动) 23.5 毫秒 | 12.0 毫秒 |
| 华南地区 | 26 | 广东广州(天翼云二区) 28.2 毫秒 | 广东汕头(4090.cn) 50.4 毫秒 | 35.9 毫秒 |
2、UCloud国内快杰云主机和国内通用型云主机基本信息对比
这里老刘博客直接开了同配置的两种类型机器，我们一起来看看国内快杰云主机和国内通用型云主机性能有什么区别。从测试结果来看，两款云服务器均选择的是Intel CPU，快杰云主机CPU主频为2.49、通用云主机CPU主频为2.19！其他的都没有什么区别。
UCloud国内快杰云主机基本信息测试结果：
-> System Information OS Release: CentOS Linux 7.6.1810 (x86_64) CPU Model: Intel Xeon Processor (Cascadelake) 2.49 GHz CPU Cache Size: 16384 KB CPU Number: 2 vCPU Virt Type: KVM Memory Usage: 194.19 MB / 3.69 GB Swap Usage: [ No Swapfile/Swap Partition ] Boot Device: /dev/vda1 Disk Usage: 1.54 GB / 20.96 GB CPU Usage: 0.0% used, 0.0% iowait, 0.0% steal Load (1/5/15min): 0.21 0.05 0.02 Uptime: 0 Days, 0 Hours, 10 Minutes, 0 Seconds Kernel Version: 4.19.0-6.el7.ucloud.x86_64 Network CC Method: cubic + pfifo_fast -> Network Infomation -> Media Unlock Test HBO Now: No Bahamut Anime: No Abema.TV: No Princess Connect Re:Dive Japan: No BBC: No BiliBili China Mainland Only: Yes BiliBili Hongkong/Macau/Taiwan: No Bilibili Taiwan Only: No
UCloud国内通用云主机基本信息测试结果：
-> System Information OS Release: CentOS Linux 7.6.1810 (x86_64) CPU Model: Intel Core Processor (Broadwell, no TSX) 2.19 GHz CPU Cache Size: 16384 KB CPU Number: 2 vCPU Virt Type: KVM Memory Usage: 179.67 MB / 3.71 GB Swap Usage: 0 KB / 512.00 MB Boot Device: /dev/vda1 Disk Usage: 2.20 GB / 20.96 GB CPU Usage: 3.3% used, 0.0% iowait, 0.0% steal Load (1/5/15min): 0.64 0.20 0.07 Uptime: 0 Days, 0 Hours, 1 Minutes, 5 Seconds Kernel Version: 3.10.0-1062.9.1.el7.x86_64 Network CC Method: cubic + pfifo_fast -> Network Infomation -> Media Unlock Test HBO Now: No Bahamut Anime: No Abema.TV: No Princess Connect Re:Dive Japan: No BBC: No BiliBili China Mainland Only: Yes BiliBili Hongkong/Macau/Taiwan: No Bilibili Taiwan Only: No
3、UCloud国内快杰云主机和通用云主机CPU、硬盘、内存性能对比
两个数据一对比就知道快杰云主机小胜那么一点点。
UCloud国内快杰云主机CPU、硬盘、内存性能结果：
-> CPU Performance Test (Standard Mode, 3-Pass @ 30sec) 1 Thread Test: 1156 Scores 2 Threads Test: 2270 Scores -> Memory Performance Test (Standard Mode, 3-Pass @ 30sec) 1 Thread - Read Test : 22424.46 MB/s 1 Thread - Write Test: 14941.51 MB/s -> Disk Speed Test (4K Block/1M Block, Direct-Write) Test Name Write Speed Read Speed 10MB-4K Block 11.2 MB/s (2725 IOPS, 0.94s) 11.3 MB/秒 (2763 IOPS, 0.93s) 10MB-1M Block 124 MB/s (118 IOPS, 0.08s) 124 MB/秒 (117 IOPS, 0.08s) 100MB-4K Block 11.4 MB/s (2773 IOPS, 9.23s) 11.3 MB/秒 (2766 IOPS, 9.25s) 100MB-1M Block 135 MB/s (128 IOPS, 0.78s) 132 MB/秒 (126 IOPS, 0.79s) 1GB-4K Block 11.4 MB/s (2784 IOPS, 91.93s) 11.4 MB/秒 (2783 IOPS, 91.98s) 1GB-1M Block 136 MB/s (129 IOPS, 7.72s) 136 MB/秒 (129 IOPS, 7.72s)
UCloud国内通用云主机CPU、硬盘、内存性能结果：
-> CPU Performance Test (Standard Mode, 3-Pass @ 30sec) 1 Thread Test: 807 Scores 2 Threads Test: 1605 Scores -> Memory Performance Test (Standard Mode, 3-Pass @ 30sec) 1 Thread - Read Test : 16672.36 MB/s 1 Thread - Write Test: 11602.70 MB/s -> Disk Speed Test (4K Block/1M Block, Direct-Write) Test Name Write Speed Read Speed 10MB-4K Block 9.5 MB/s (2324 IOPS, 1.10s) 53.6 MB/秒 (13077 IOPS, 0.20s) 10MB-1M Block 50.4 MB/s (48 IOPS, 0.21s) 627 MB/秒 (597 IOPS, 0.02s) 100MB-4K Block 14.4 MB/s (3512 IOPS, 7.29s) 59.9 MB/秒 (14634 IOPS, 1.75s) 100MB-1M Block 124 MB/s (118 IOPS, 0.84s) 1.1 GB/秒 (1074 IOPS, 0.09s) 1GB-4K Block 10.4 MB/s (2533 IOPS, 101.05s) 65.1 MB/秒 (15883 IOPS, 16.12s) 1GB-1M Block 47.8 MB/s (45 IOPS, 21.95s) 1.1 GB/秒 (1052 IOPS, 0.95s)
4、UCloud国内快杰云主机和通用云主机UnixBench跑分测试
快杰云主机得分2881.5、通用云主机得分1640.3，从跑分来看优势还是很明显的。
UCloud国内快杰云主机UnixBench跑分测试结果：
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: 10-25-133-159: GNU/Linux
OS: GNU/Linux -- 4.19.0-6.el7.ucloud.x86_64 -- #1 SMP Wed Feb 12 07:32:16 UTC 2020
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel Xeon Processor (Cascadelake) (4988.3 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: Intel Xeon Processor (Cascadelake) (4988.3 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
16:55:42 up 42 min, 4 users, load average: 0.08, 0.04, 0.16; runlevel 3
------------------------------------------------------------------------
Benchmark Run: 四 7月 16 2020 16:55:42 - 17:24:02
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 42165292.6 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4268.9 MWIPS (11.4 s, 7 samples)
Execl Throughput 4750.6 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1256432.4 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 354527.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 3275978.6 KBps (30.0 s, 2 samples)
Pipe Throughput 2093002.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching 323729.3 lps (10.0 s, 7 samples)
Process Creation 9865.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 8829.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1582.9 lpm (60.0 s, 2 samples)
System Call Overhead 2693993.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 42165292.6 3613.1
Double-Precision Whetstone 55.0 4268.9 776.2
Execl Throughput 43.0 4750.6 1104.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 1256432.4 3172.8
File Copy 256 bufsize 500 maxblocks 1655.0 354527.4 2142.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 3275978.6 5648.2
Pipe Throughput 12440.0 2093002.3 1682.5
Pipe-based Context Switching 4000.0 323729.3 809.3
Process Creation 126.0 9865.5 783.0
Shell Scripts (1 concurrent) 42.4 8829.9 2082.5
Shell Scripts (8 concurrent) 6.0 1582.9 2638.2
System Call Overhead 15000.0 2693993.4 1796.0
========
System Benchmarks Index Score 1811.8
------------------------------------------------------------------------
Benchmark Run: 四 7月 16 2020 17:24:02 - 17:52:09
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 77538758.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 9379.8 MWIPS (9.9 s, 7 samples)
Execl Throughput 9126.8 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1523197.9 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 405813.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 4451113.4 KBps (30.0 s, 2 samples)
Pipe Throughput 4144281.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 684772.7 lps (10.0 s, 7 samples)
Process Creation 18845.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 12802.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1828.4 lpm (60.0 s, 2 samples)
System Call Overhead 3597754.1 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 77538758.8 6644.3
Double-Precision Whetstone 55.0 9379.8 1705.4
Execl Throughput 43.0 9126.8 2122.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 1523197.9 3846.5
File Copy 256 bufsize 500 maxblocks 1655.0 405813.1 2452.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 4451113.4 7674.3
Pipe Throughput 12440.0 4144281.6 3331.4
Pipe-based Context Switching 4000.0 684772.7 1711.9
Process Creation 126.0 18845.0 1495.6
Shell Scripts (1 concurrent) 42.4 12802.4 3019.4
Shell Scripts (8 concurrent) 6.0 1828.4 3047.3
System Call Overhead 15000.0 3597754.1 2398.5
========
System Benchmarks Index Score 2881.5
======= Script description and score comparison completed! =======
UCloud国内通用云主机UnixBench跑分测试结果：
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: 10-25-70-182: GNU/Linux
OS: GNU/Linux -- 3.10.0-1062.9.1.el7.x86_64 -- #1 SMP Fri Dec 6 15:49:49 UTC 2019
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel Core Processor (Broadwell, no TSX) (4389.8 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: Intel Core Processor (Broadwell, no TSX) (4389.8 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
16:38:05 up 12 min, 2 users, load average: 0.60, 0.78, 0.52; runlevel 3
------------------------------------------------------------------------
Benchmark Run: 四 7月 16 2020 16:38:05 - 17:06:08
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 29327469.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4430.7 MWIPS (9.8 s, 7 samples)
Execl Throughput 2226.0 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 541736.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 141915.6 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1650426.7 KBps (30.0 s, 2 samples)
Pipe Throughput 753025.5 lps (10.0 s, 7 samples)
Pipe-based Context Switching 122447.1 lps (10.0 s, 7 samples)
Process Creation 6495.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 5167.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1055.4 lpm (60.0 s, 2 samples)
System Call Overhead 737533.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 29327469.2 2513.1
Double-Precision Whetstone 55.0 4430.7 805.6
Execl Throughput 43.0 2226.0 517.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 541736.0 1368.0
File Copy 256 bufsize 500 maxblocks 1655.0 141915.6 857.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 1650426.7 2845.6
Pipe Throughput 12440.0 753025.5 605.3
Pipe-based Context Switching 4000.0 122447.1 306.1
Process Creation 126.0 6495.6 515.5
Shell Scripts (1 concurrent) 42.4 5167.4 1218.7
Shell Scripts (8 concurrent) 6.0 1055.4 1759.0
System Call Overhead 15000.0 737533.6 491.7
========
System Benchmarks Index Score 916.7
------------------------------------------------------------------------
Benchmark Run: 四 7月 16 2020 17:06:08 - 17:34:12
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 58010210.0 lps (10.0 s, 7 samples)
Double-Precision Whetstone 8879.0 MWIPS (9.8 s, 7 samples)
Execl Throughput 4707.3 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 953632.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 250958.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2859529.2 KBps (30.0 s, 2 samples)
Pipe Throughput 1478478.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 243990.1 lps (10.0 s, 7 samples)
Process Creation 14524.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 7811.0 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1066.3 lpm (60.1 s, 2 samples)
System Call Overhead 1315582.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 58010210.0 4970.9
Double-Precision Whetstone 55.0 8879.0 1614.4
Execl Throughput 43.0 4707.3 1094.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 953632.0 2408.2
File Copy 256 bufsize 500 maxblocks 1655.0 250958.8 1516.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 2859529.2 4930.2
Pipe Throughput 12440.0 1478478.1 1188.5
Pipe-based Context Switching 4000.0 243990.1 610.0
Process Creation 126.0 14524.5 1152.7
Shell Scripts (1 concurrent) 42.4 7811.0 1842.2
Shell Scripts (8 concurrent) 6.0 1066.3 1777.1
System Call Overhead 15000.0 1315582.4 877.1
========
System Benchmarks Index Score 1640.3
======= Script description and score comparison completed! =======
5、UCloud国内快杰云主机和通用云主机宝塔跑分测试
快杰云主机宝塔跑分11566，通用云主机宝塔跑分11481，没有明显区别。
快杰云主机宝塔跑分测试结果：
通用云主机宝塔跑分测试结果：
6、UCloud国内快杰云主机速度测试
国内来说是没有问题的，上行5M可以跑满，下能跑个70-80Mbps的样子。
-> Speedtest.net Network Speed Test Node Name Upload Speed Download Speed Ping Latency Speedtest Default 0.57 MB/s 6.08 MB/s 17.94 ms China, Jilin CU 0.57 MB/s 5.66 MB/s 41.91 ms China, Shandong CU 0.57 MB/s 7.54 MB/s 30.70 ms China, Nanjing CU 0.57 MB/s 5.74 MB/s 7.44 ms China, Shanghai CU 0.57 MB/s 5.74 MB/s 2.02 ms China, Lanzhou CU 0.57 MB/s 6.14 MB/s 35.58 ms China, Hangzhou CT 0.57 MB/s 5.75 MB/s 9.82 ms China, Nanjing CT 0.57 MB/s 5.70 MB/s 9.36 ms China, Guangzhou CT 0.57 MB/s 5.46 MB/s 27.96 ms China, Wuhan CT 0.57 MB/s 5.71 MB/s 17.63 ms China, Shenyang CM 0.57 MB/s 6.77 MB/s 41.78 ms China, Nanning CM 0.57 MB/s 7.89 MB/s 54.43 ms China, Lanzhou CM 0.57 MB/s 5.64 MB/s 46.83 ms
好啦，以上就是老刘博客总结的UCloud国内快杰云主机和通用云主机性能评测对比全部内容，关于快杰云主机高性能的原理，想必不少朋友想知道，那么你可以查阅老刘博客前文《UCloud优刻得快杰云主机原理大揭秘》。
转载自https://www.zhuji999.com/1146.html
