Azure香港机器性能测试-A2机型(Intel E5系列CPU)

Azure的A系列机器默认采用的是AMD系列的CPU，有些人不喜欢，我们可以通过一些手段开一个Intel CPU的A系列机器，这个方法就是先开通一个D系列的机器，这样就得到了一个Intel E5 CPU的机器，然后再将降级为A系列的机器，这样我们就得到了一个Intel CPU的A系列机器，但是貌似性能和AMD的没区别，甚至还跑分还没有AMD的高，不过对于Intel控来说也可以尝试一下。下面是通过上述方法得到的一个A2机器的测试情况：

1. bench.sh测试（测试工具使用参考这里）
   • 
     

     CPU model	Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
     Number of cores	2
     CPU frequency	2199.985 MHz
     Total amount of ram	3440 MB
     Total amount of swap	0 MB
     System uptime	0days, 0:19:41

   • Node Name	Node IP address	Download Speed
     CacheFly	205.234.175.175	16.6MB/s
     Linode, Tokyo, JP	106.187.96.148	37.0MB/s
     Linode, Singapore, SG	139.162.23.4	28.0MB/s
     Linode, London, UK	176.58.107.39	6.69MB/s
     Linode, Frankfurt, DE	139.162.130.8	3.25MB/s
     Linode, Fremont, CA	50.116.14.9	7.29MB/s
     Vultr, Dallas, TX	108.61.224.175	7.93MB/s
     Vultr, Seattle, WA	108.61.194.105	8.73MB/s
     Vultr, Los Angeles, CA	108.61.219.200	9.52MB/s
     Vultr, Frankfurt, DE	108.61.210.117	4.19MB/s
     Vultr, Tokyo, JP	108.61.201.151	32.3MB/s
     Softlayer, Dallas, TX	173.192.68.18	10.9MB/s
     Softlayer, Seattle, WA	67.228.112.250	16.4MB/s
     Softlayer, Frankfurt, DE	159.122.69.4	8.10MB/s
     Softlayer, Singapore, SG	119.81.28.170	16.1MB/s
     Softlayer, HongKong, CN	119.81.130.170	95.1MB/s

   • I/O speed : 1073741824 bytes (1.1 GB) copied, 51.1787 s, 21.0 MB/s
2. unixbench测试（测试工具使用参考这里）
   • 单核性能测试
     

     Dhrystone 2 using register variables	14581516.2 lps   (10.0 s, 7 samples)
     Double-Precision Whetstone	1743.1 MWIPS (9.7 s, 7 samples)
     Execl Throughput	2447.4 lps   (29.8 s, 2 samples)
     File Copy 1024 bufsize 2000 maxblocks	431340.4 KBps  (30.0 s, 2 samples)
     File Copy 256 bufsize 500 maxblocks	120693.5 KBps  (30.0 s, 2 samples)
     File Copy 4096 bufsize 8000 maxblocks	1004188.2 KBps  (30.0 s, 2 samples)
     Pipe Throughput	782700.2 lps   (10.0 s, 7 samples)
     Pipe-based Context Switching	157933.9 lps   (10.0 s, 7 samples)
     Process Creation	5029.5 lps   (30.0 s, 2 samples)
     Shell Scripts (1 concurrent)	5024.5 lpm   (60.0 s, 2 samples)
     Shell Scripts (8 concurrent)	873.1 lpm   (60.0 s, 2 samples)
     System Call Overhead	1706423.6 lps   (10.0 s, 7 samples)

   • System Benchmarks Index Values	BASELINE	RESULT	INDEX
     Dhrystone 2 using register variables	116700	14581516.2	1249.5
     Double-Precision Whetstone	55	1743.1	316.9
     Execl Throughput	43	2447.4	569.2
     File Copy 1024 bufsize 2000 maxblocks	3960	431340.4	1089.2
     File Copy 256 bufsize 500 maxblocks	1655	120693.5	729.3
     File Copy 4096 bufsize 8000 maxblocks	5800	1004188.2	1731.4
     Pipe Throughput	12440	782700.2	629.2
     Pipe-based Context Switching	4000	157933.9	394.8
     Process Creation	126	5029.5	399.2
     Shell Scripts (1 concurrent)	42.4	5024.5	1185
     Shell Scripts (8 concurrent)	6	873.1	1455.1
     System Call Overhead	15000	1706423.6	1137.6

   • System Benchmarks Index Score：791.3
   • 双核性能测试

   • Dhrystone 2 using register variables	29146257.2 lps   (10.0 s, 7 samples)
     Double-Precision Whetstone	3501.5 MWIPS (9.8 s, 7 samples)
     Execl Throughput	4054.0 lps   (29.9 s, 2 samples)
     File Copy 1024 bufsize 2000 maxblocks	538224.0 KBps  (30.0 s, 2 samples)
     File Copy 256 bufsize 500 maxblocks	150159.8 KBps  (30.0 s, 2 samples)
     File Copy 4096 bufsize 8000 maxblocks	1340262.9 KBps  (30.0 s, 2 samples)
     Pipe Throughput	1569056.4 lps   (10.0 s, 7 samples)
     Pipe-based Context Switching	311207.3 lps   (10.0 s, 7 samples)
     Process Creation	8196.4 lps   (30.0 s, 2 samples)
     Shell Scripts (1 concurrent)	6105.8 lpm   (60.0 s, 2 samples)
     Shell Scripts (8 concurrent)	947.5 lpm   (60.0 s, 2 samples)
     System Call Overhead	2810230.6 lps   (10.0 s, 7 samples)

   • System Benchmarks Index Values	BASELINE	RESULT	INDEX
     Dhrystone 2 using register variables	116700	29146257.2	2497.5
     Double-Precision Whetstone	55	3501.5	636.6
     Execl Throughput	43	4054	942.8
     File Copy 1024 bufsize 2000 maxblocks	3960	538224	1359.2
     File Copy 256 bufsize 500 maxblocks	1655	150159.8	907.3
     File Copy 4096 bufsize 8000 maxblocks	5800	1340262.9	2310.8
     Pipe Throughput	12440	1569056.4	1261.3
     Pipe-based Context Switching	4000	311207.3	778
     Process Creation	126	8196.4	650.5
     Shell Scripts (1 concurrent)	42.4	6105.8	1440
     Shell Scripts (8 concurrent)	6	947.5	1579.2
     System Call Overhead	15000	2810230.6	1873.5

   • System Benchmarks Index Score：1226.8