老刘测评
UCloud快杰云主机怎么样？UCloud优刻得基于快杰云主机先后推出了第一代Intel CPU版本和第二代AMD CPU版本，老刘测评先前介绍和测评的均为Intel快杰云主机（毕竟人家厂商最近促销的是这个版本嘛），那么到底快杰云主机Intel和AMD版哪个好呢？老刘测评选择了UCloud上海二A可用区，配置是8核16G内存10M带宽系统盘40G的上海VPS进行性能测评对比，目前该配置的机器UCloud促销价是7888元一年>>
和前文《优刻得1C2G2M香港通用型与快杰型云主机性能测评对比》一样，此次老刘继续使用的测试工具是站长工具Ping值测试，SuperBench脚本测试和UnixBench脚本跑分。
Ping是潜水艇人员的专用术语，表示回应的声纳脉冲，在网络中Ping是一个十分好用的TCP/IP工具。它主要的功能是用来检测网络的连通情况和分析网络速度。
SuperBench脚本，可以一键测试云主机服务器VPS各种性能，包括机器基本信息，硬盘读写速度，到国内各个地区的下载速度等。
UnixBench是一个类unix系（Unix，BSD，Linux）统下的性能测试工具，一个开源工具，被广泛用与测试linux系统主机的性能。Unixbench的主要测试项目有：系统调用、读写、进程、图形化测试、2D、3D、管道、运算、C库等系统基准性能提供测试数据，UnixBench运行结束后会给出机器整体性能跑分值，分数越高机器性能越好，一般跑分在1200-1300以上的机器，性能不错，值得入手。
Ping值测试对比
老刘使用站长工具Ping测试分别对Intel和AMD版的快杰云主机进行了ping值测试，以下是测试结果图：
Intel快杰云主机Ping值
AMD快杰云主机Ping值
分析：由以上两图可以看出，Intel和AMD版的两种CPU快杰云主机，在电信、多线、联通和移动线路上响应时间一致，平均值都是27ms，超低延时没得说！
SuperBench测速对比
Intel快杰云主机SuperBench测试结果
----------------------------------------------------------------------
Superbench.sh -- https://www.oldking.net/350.html
Mode : Standard Version : 1.1.7
Usage : wget -qO- sb.oldking.net | bash
----------------------------------------------------------------------
CPU Model : Intel Xeon Processor (Cascadelake)
CPU Cores : 8 Cores 2494.140 MHz x86_64
CPU Cache : 16384 KB
OS : CentOS 7.6.1810 (64 Bit) KVM
Kernel : 4.19.0-6.el7.ucloud.x86_64
Total Space : 1.4 GB / 40.0 GB
Total RAM : 146 MB / 15814 MB (340 MB Buff)
Total SWAP : 0 MB / 0 MB
Uptime : 0 days 0 hour 6 min
Load Average : 0.16, 0.14, 0.05
TCP CC : cubic
ASN & ISP : AS4812, China Telecom (Group)
Organization : Shanghai UCloud Information Technology Company Limited
Location : Beijing, China / CN
Region : Beijing
----------------------------------------------------------------------
I/O Speed( 1.0GB ) : 139 MB/s
I/O Speed( 1.0GB ) : 137 MB/s
I/O Speed( 1.0GB ) : 139 MB/s
Average I/O Speed : 138.3 MB/s
----------------------------------------------------------------------
Node Name Upload Speed Download Speed Latency
Speedtest.net 9.54 Mbit/s 47.85 Mbit/s 15.69 ms
Fast.com 0.00 Mbit/s 36.6 Mbit/s -
Nanjing 5G CT 9.54 Mbit/s 47.79 Mbit/s 8.26 ms
Hefei 5G CT 9.54 Mbit/s 47.70 Mbit/s 10.99 ms
Guangzhou 5G CT 9.53 Mbit/s 41.93 Mbit/s 30.44 ms
TianJin 5G CU 9.62 Mbit/s 48.43 Mbit/s 20.36 ms
Shanghai 5G CU 9.62 Mbit/s 48.30 Mbit/s 2.10 ms
Guangzhou 5G CU 9.54 Mbit/s 52.96 Mbit/s 29.10 ms
Tianjin 5G CM 9.66 Mbit/s 50.41 Mbit/s 33.26 ms
Wuxi 5G CM 9.63 Mbit/s 48.20 Mbit/s 11.14 ms
Nanjing 5G CM 9.56 Mbit/s 37.15 Mbit/s 14.35 ms
Hefei 5G CM 9.55 Mbit/s 54.76 Mbit/s 19.20 ms
Changsha 5G CM 9.63 Mbit/s 48.37 Mbit/s 26.96 ms
----------------------------------------------------------------------
Finished in : 8 min 28 sec
Timestamp : 2020-07-29 18:44:21 GMT+8
Results : ./superbench.log
----------------------------------------------------------------------
AMD快杰云主机SuperBench测试结果
----------------------------------------------------------------------
Superbench.sh -- https://www.oldking.net/350.html
Mode : Standard Version : 1.1.7
Usage : wget -qO- sb.oldking.net | bash
----------------------------------------------------------------------
CPU Model : AMD EPYC Processor
CPU Cores : 8 Cores 2894.534 MHz x86_64
CPU Cache : 512 KB
OS : CentOS 7.6.1810 (64 Bit) KVM
Kernel : 4.19.0-6.el7.ucloud.x86_64
Total Space : 1.4 GB / 40.0 GB
Total RAM : 143 MB / 15814 MB (337 MB Buff)
Total SWAP : 0 MB / 0 MB
Uptime : 0 days 0 hour 3 min
Load Average : 0.01, 0.03, 0.00
TCP CC : cubic
ASN & ISP : AS4812, China Telecom (Group)
Organization : Shanghai UCloud Information Technology Company Limited
Location : Beijing, China / CN
Region : Beijing
----------------------------------------------------------------------
I/O Speed( 1.0GB ) : 135 MB/s
I/O Speed( 1.0GB ) : 136 MB/s
I/O Speed( 1.0GB ) : 136 MB/s
Average I/O Speed : 135.7 MB/s
----------------------------------------------------------------------
Node Name Upload Speed Download Speed Latency
Speedtest.net 9.57 Mbit/s 50.80 Mbit/s 17.41 ms
Fast.com 0.00 Mbit/s 7.0 Mbit/s -
Nanjing 5G CT 9.54 Mbit/s 47.87 Mbit/s 9.43 ms
Hefei 5G CT 9.54 Mbit/s 47.82 Mbit/s 10.36 ms
Guangzhou 5G CT 9.54 Mbit/s 43.24 Mbit/s 27.55 ms
TianJin 5G CU 9.63 Mbit/s 48.24 Mbit/s 20.51 ms
Shanghai 5G CU 9.62 Mbit/s 48.28 Mbit/s 2.13 ms
Guangzhou 5G CU 9.54 Mbit/s 48.25 Mbit/s 28.83 ms
Tianjin 5G CM 9.64 Mbit/s 68.29 Mbit/s 33.25 ms
Wuxi 5G CM 9.63 Mbit/s 48.18 Mbit/s 10.65 ms
Nanjing 5G CM 9.52 Mbit/s 26.47 Mbit/s 12.68 ms
Hefei 5G CM 9.55 Mbit/s 52.17 Mbit/s 19.66 ms
Changsha 5G CM 9.62 Mbit/s 48.35 Mbit/s 27.83 ms
----------------------------------------------------------------------
Finished in : 8 min 46 sec
Timestamp : 2020-07-29 18:53:06 GMT+8
Results : ./superbench.log
----------------------------------------------------------------------
分析：在基本信息上可以看到，Intel快杰云主机搭载的CPU是Intel Xeon Processor，AMD快杰云主机搭载的CPU是AMD EPYC Processor；前者的CPU缓存明显高于后者，另外RAM略有区别，不过整体一致，机器配置如官方描述。Intel快杰云主机的I/O读写速度在138.3MB/s，而AMD快杰云主机在135.7MB/s，Intel快杰云主机在I/O磁盘读写性能方面更好一些，不过基本可以忽略；测速国内节点，Intel和AMD cpu快杰云主机基本一致，上行下行均能跑满，延时也无差别，均是超低延时。
UnixBench跑分对比
Intel快杰云主机跑分结果
------------------------------------------------------------------------
Benchmark Run: 三 7月 29 2020 20:05:54 - 20:34:08
8 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 41463079.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4524.7 MWIPS (10.5 s, 7 samples)
Execl Throughput 5434.6 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1284696.2 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 346429.6 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 3282364.7 KBps (30.0 s, 2 samples)
Pipe Throughput 2112918.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching 287838.1 lps (10.0 s, 7 samples)
Process Creation 7793.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 8103.3 lpm (60.0 s, 2
samples)
Shell Scripts (8 concurrent) 3275.9 lpm (60.0 s, 2 samples)
System Call Overhead 2622878.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 41463079.5 3553.0
Double-Precision Whetstone 55.0 4524.7 822.7
Execl Throughput 43.0 5434.6 1263.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 1284696.2 3244.2
File Copy 256 bufsize 500 maxblocks 1655.0 346429.6 2093.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 3282364.7 5659.2
Pipe Throughput 12440.0 2112918.3 1698.5
Pipe-based Context Switching 4000.0 287838.1 719.6
Process Creation 126.0 7793.7 618.5
Shell Scripts (1 concurrent) 42.4 8103.3 1911.2
Shell Scripts (8 concurrent) 6.0 3275.9 5459.8
System Call Overhead 15000.0 2622878.5 1748.6
========
System Benchmarks Index Score 1880.7
------------------------------------------------------------------------
Benchmark Run: 三 7月 29 2020 20:34:08 - 21:02:25
8 CPUs in system; running 8 parallel copies of tests
Dhrystone 2 using register variables 327014239.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 36663.7 MWIPS (10.3 s, 7 samples)
Execl Throughput 23798.4 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 537406.6 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 150243.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1857587.1 KBps (30.0 s, 2 samples)
Pipe Throughput 16400188.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching 2454551.4 lps (10.0 s, 7 samples)
Process Creation 37526.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 33227.1 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 4898.6 lpm (60.0 s, 2 samples)
System Call Overhead 3693549.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 327014239.5 28021.8
Double-Precision Whetstone 55.0 36663.7 6666.1
Execl Throughput 43.0 23798.4 5534.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 537406.6 1357.1
File Copy 256 bufsize 500 maxblocks 1655.0 150243.4 907.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 1857587.1 3202.7
Pipe Throughput 12440.0 16400188.7 13183.4
Pipe-based Context Switching 4000.0 2454551.4 6136.4
Process Creation 126.0 37526.7 2978.3
Shell Scripts (1 concurrent) 42.4 33227.1 7836.6
Shell Scripts (8 concurrent) 6.0 4898.6 8164.3
System Call Overhead 15000.0 3693549.5 2462.4
========
System Benchmarks Index Score 4813.8
# ========================================================= #
# Script description and score comparison completed!
# For more information, please go to https://unixbench.org
# ========================================================= #
AMD快杰云主机跑分结果
------------------------------------------------------------------------
Benchmark Run: 三 7月 29 2020 20:08:29 - 20:38:05
8 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 39879697.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3019.1 MWIPS (19.7 s, 7 samples)
Execl Throughput 5969.2 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1174269.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 333552.6 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 3221619.1 KBps (30.0 s, 2 samples)
Pipe Throughput 2157534.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 259542.4 lps (10.0 s, 7 samples)
Process Creation 6929.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 8371.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 3161.3 lpm (60.0 s, 2 samples)
System Call Overhead 2626090.2 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 39879697.4 3417.3
Double-Precision Whetstone 55.0 3019.1 548.9
Execl Throughput 43.0 5969.2 1388.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 1174269.1 2965.3
File Copy 256 bufsize 500 maxblocks 1655.0 333552.6 2015.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 3221619.1 5554.5
Pipe Throughput 12440.0 2157534.6 1734.4
Pipe-based Context Switching 4000.0 259542.4 648.9
Process Creation 126.0 6929.0 549.9
Shell Scripts (1 concurrent) 42.4 8371.6 1974.4
Shell Scripts (8 concurrent) 6.0 3161.3 5268.8
System Call Overhead 15000.0 2626090.2 1750.7
========
System Benchmarks Index Score 1774.4
------------------------------------------------------------------------
Benchmark Run: 三 7月 29 2020 20:38:05 - 21:07:38
8 CPUs in system; running 8 parallel copies of tests
Dhrystone 2 using register variables 315079880.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 26573.0 MWIPS (17.1 s, 7 samples)
Execl Throughput 20635.8 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 595697.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 151408.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2190231.0 KBps (30.0 s, 2 samples)
Pipe Throughput 15611882.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 2321537.8 lps (10.0 s, 7 samples)
Process Creation 32487.9 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 31452.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 4701.1 lpm (60.0 s, 2 samples)
System Call Overhead 4922452.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 315079880.8 26999.1
Double-Precision Whetstone 55.0 26573.0 4831.4
Execl Throughput 43.0 20635.8 4799.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 595697.1 1504.3
File Copy 256 bufsize 500 maxblocks 1655.0 151408.8 914.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 2190231.0 3776.3
Pipe Throughput 12440.0 15611882.4 12549.7
Pipe-based Context Switching 4000.0 2321537.8 5803.8
Process Creation 126.0 32487.9 2578.4
Shell Scripts (1 concurrent) 42.4 31452.4 7418.0
Shell Scripts (8 concurrent) 6.0 4701.1 7835.1
System Call Overhead 15000.0 4922452.4 3281.6
========
System Benchmarks Index Score 4701.1
# ========================================================= #
# Script description and score comparison completed!
# For more information, please go to https://unixbench.org
# ========================================================= #
分析：UnixBench测试了系统调用、读写、进程、图形化测试、2D、3D、管道、运算、C库等性能，是一个比较综合性测评机器性能的工具，这里具体各个性能数据不用作分析，这里我们只需要看最后的跑分结果。Intel快杰云主机单线程跑分1880.7，8线程（CPU使用率100%时）跑分4813.8；Intel快杰云主机单线程跑分1774.4，8线程跑分而快杰型云主机跑分4701.1。无论单线程，8线程，跑分均是在1200-1300分值以上，性能优秀，同时Intel快杰云主机跑分更高，综合性能较AMD快杰云主机更好。
总结，老刘测评通过以上测试结果分析可知，UCloud优刻得上海VPS云主机8核16G配置的Intel和AMD快杰云主机在连通性和速度上基本无差别，Intel版本较AMD版本在磁盘读写性能上略微好一些，同时在云主机综合性能上，Intel版本更胜一筹，我想这也是为什么Intel版本快杰云主机为什么比AMD版本的快杰云主机价格更贵一些的原因吧~
