SA1的EPYC 7551和普通型号的Xeon Gold 6133/6148比,IPC相差是有四分之一,大致是E5-2695 v2水平
但VPS肯定不会给你完整的性能,实际的效能取决于服务商的资源限制,在腾讯这里,二者给的基准性能大致相当
对性能和指令集没有特别要求的其实有可以选SA1就好了,价钱便宜不少
https://www.cpubenchmark.net/compare/AMD-EPYC-7551-vs-Intel-Xeon-Gold-6148-vs-Intel-Xeon-E5-2695-v2/3089vs3176vs2114
SA1:AMD EPYC 7551 32-Core Processor @ 2.0GHz
https://serverscope.io/trials/29Vj
- BYTE UNIX Benchmarks (Version 5.1.3)
- System: VM-0-17-ubuntu: GNU/Linux
- OS: GNU/Linux -- 4.15.0-29-generic -- #31-Ubuntu SMP Tue Jul 17 15:39:52 UTC 2018
- Machine: x86_64 (x86_64)
- Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
- CPU 0: AMD EPYC 7551 32-Core Processor (4000.0 bogomips)
- x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
- 00:33:17 up 11 min, 1 user, load average: 7.11, 4.84, 2.17; runlevel 2019-06-03
- ------------------------------------------------------------------------
- Benchmark Run: Mon Jun 03 2019 00:33:17 - 01:01:20
- 1 CPU in system; running 1 parallel copy of tests
- Dhrystone 2 using register variables 30102533.3 lps (10.0 s, 7 samples)
- Double-Precision Whetstone 4631.3 MWIPS (9.6 s, 7 samples)
- Execl Throughput 3893.8 lps (30.0 s, 2 samples)
- File Copy 1024 bufsize 2000 maxblocks 655289.5 KBps (30.0 s, 2 samples)
- File Copy 256 bufsize 500 maxblocks 183380.2 KBps (30.0 s, 2 samples)
- File Copy 4096 bufsize 8000 maxblocks 1701823.2 KBps (30.0 s, 2 samples)
- Pipe Throughput 1191447.5 lps (10.0 s, 7 samples)
- Pipe-based Context Switching 196770.2 lps (10.0 s, 7 samples)
- Process Creation 13257.3 lps (30.0 s, 2 samples)
- Shell Scripts (1 concurrent) 7216.6 lpm (60.0 s, 2 samples)
- Shell Scripts (8 concurrent) 908.1 lpm (60.0 s, 2 samples)
- System Call Overhead 1889844.1 lps (10.0 s, 7 samples)
- System Benchmarks Index Values BASELINE RESULT INDEX
- Dhrystone 2 using register variables 116700.0 30102533.3 2579.5
- Double-Precision Whetstone 55.0 4631.3 842.1
- Execl Throughput 43.0 3893.8 905.5
- File Copy 1024 bufsize 2000 maxblocks 3960.0 655289.5 1654.8
- File Copy 256 bufsize 500 maxblocks 1655.0 183380.2 1108.0
- File Copy 4096 bufsize 8000 maxblocks 5800.0 1701823.2 2934.2
- Pipe Throughput 12440.0 1191447.5 957.8
- Pipe-based Context Switching 4000.0 196770.2 491.9
- Process Creation 126.0 13257.3 1052.2
- Shell Scripts (1 concurrent) 42.4 7216.6 1702.0
- Shell Scripts (8 concurrent) 6.0 908.1 1513.5
- System Call Overhead 15000.0 1889844.1 1259.9
- ========
- System Benchmarks Index Score 1266.5
复制代码
S4:Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz
https://serverscope.io/trials/q76n
- BYTE UNIX Benchmarks (Version 5.1.3)
- System: Skylake-demo: GNU/Linux
- OS: GNU/Linux -- 4.15.0-29-generic -- #31-Ubuntu SMP Tue Jul 17 15:39:52 UTC 2018
- Machine: x86_64 (x86_64)
- Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
- CPU 0: Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz (4788.7 bogomips)
- x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
- 00:33:42 up 12 min, 1 user, load average: 8.59, 4.90, 2.11; runlevel 2019-06-03
- ------------------------------------------------------------------------
- Benchmark Run: Mon Jun 03 2019 00:33:42 - 01:01:44
- 1 CPU in system; running 1 parallel copy of tests
- Dhrystone 2 using register variables 38939560.2 lps (10.0 s, 7 samples)
- Double-Precision Whetstone 4783.2 MWIPS (9.1 s, 7 samples)
- Execl Throughput 5396.8 lps (30.0 s, 2 samples)
- File Copy 1024 bufsize 2000 maxblocks 751051.1 KBps (30.0 s, 2 samples)
- File Copy 256 bufsize 500 maxblocks 203897.7 KBps (30.0 s, 2 samples)
- File Copy 4096 bufsize 8000 maxblocks 1879795.5 KBps (30.0 s, 2 samples)
- Pipe Throughput 1111914.8 lps (10.0 s, 7 samples)
- Pipe-based Context Switching 248750.9 lps (10.0 s, 7 samples)
- Process Creation 15632.8 lps (30.0 s, 2 samples)
- Shell Scripts (1 concurrent) 10098.0 lpm (60.0 s, 2 samples)
- Shell Scripts (8 concurrent) 1299.0 lpm (60.0 s, 2 samples)
- System Call Overhead 820794.3 lps (10.0 s, 7 samples)
- System Benchmarks Index Values BASELINE RESULT INDEX
- Dhrystone 2 using register variables 116700.0 38939560.2 3336.7
- Double-Precision Whetstone 55.0 4783.2 869.7
- Execl Throughput 43.0 5396.8 1255.1
- File Copy 1024 bufsize 2000 maxblocks 3960.0 751051.1 1896.6
- File Copy 256 bufsize 500 maxblocks 1655.0 203897.7 1232.0
- File Copy 4096 bufsize 8000 maxblocks 5800.0 1879795.5 3241.0
- Pipe Throughput 12440.0 1111914.8 893.8
- Pipe-based Context Switching 4000.0 248750.9 621.9
- Process Creation 126.0 15632.8 1240.7
- Shell Scripts (1 concurrent) 42.4 10098.0 2381.6
- Shell Scripts (8 concurrent) 6.0 1299.0 2165.0
- System Call Overhead 15000.0 820794.3 547.2
- ========
- System Benchmarks Index Score 1393.6
复制代码 |