全球主机交流论坛

标题: SunnyVision家AMD4000 UnixBench跑分，大佬们传个intel的 [打印本页]

作者: iamyg 时间: 2018-1-18 10:19
标题: SunnyVision家AMD4000 UnixBench跑分，大佬们传个intel的
本帖最后由 iamyg 于 2018-1-19 10:31 编辑

更新：intel L5630跑分比AMD4000这个高了1000分，见后续大佬发的分数。新购买了个1240v1，跑分是4600，见下面

------------------------------------------------------------------------
Benchmark Run: Thu Jan 18 2018 09:09:38 - 09:37:36
12 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables    15569799.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone                   2017.3 MWIPS (9.9 s, 7 samples)
Execl Throughput                            1256.9 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       477631.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks       138144.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       789006.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                            840979.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching                50212.1 lps (10.0 s, 7 samples)
Process Creation                            1999.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                1785.1 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   872.9 lpm (60.0 s, 2 samples)
System Call Overhead                      943247.0 lps (10.0 s, 7 samples)

System Benchmarks Index Values             BASELINE    RESULT INDEX
Dhrystone 2 using register variables       116700.0 15569799.4 1334.2
Double-Precision Whetstone                      55.0    2017.3 366.8
Execl Throughput                               43.0    1256.9 292.3
File Copy 1024 bufsize 2000 maxblocks       3960.0    477631.8 1206.1
File Copy 256 bufsize 500 maxblocks          1655.0    138144.3 834.7
File Copy 4096 bufsize 8000 maxblocks       5800.0    789006.0 1360.4
Pipe Throughput                            12440.0    840979.1 676.0
Pipe-based Context Switching                4000.0    50212.1 125.5
Process Creation                               126.0    1999.4 158.7
Shell Scripts (1 concurrent)                   42.4    1785.1 421.0
Shell Scripts (8 concurrent)                   6.0       872.9 1454.9
System Call Overhead                         15000.0    943247.0 628.8
                                                               ========
System Benchmarks Index Score                                        563.1

------------------------------------------------------------------------
Benchmark Run: Thu Jan 18 2018 09:37:36 - 10:05:39
12 CPUs in system; running 12 parallel copies of tests

Dhrystone 2 using register variables    186949499.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone                   24146.0 MWIPS (9.9 s, 7 samples)
Execl Throughput                            18063.2 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       511424.1 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks       154866.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks    1085746.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                         10104303.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching             1398551.2 lps (10.0 s, 7 samples)
Process Creation                            32470.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                22965.7 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                3668.1 lpm (60.1 s, 2 samples)
System Call Overhead                      1789273.3 lps (10.0 s, 7 samples)

System Benchmarks Index Values             BASELINE    RESULT INDEX
Dhrystone 2 using register variables       116700.0  186949499.2  16019.7
Double-Precision Whetstone                      55.0    24146.0 4390.2
Execl Throughput                               43.0    18063.2 4200.7
File Copy 1024 bufsize 2000 maxblocks       3960.0    511424.1 1291.5
File Copy 256 bufsize 500 maxblocks          1655.0    154866.0 935.7
File Copy 4096 bufsize 8000 maxblocks       5800.0 1085746.0 1872.0
Pipe Throughput                            12440.0 10104303.0 8122.4
Pipe-based Context Switching                4000.0 1398551.2 3496.4
Process Creation                               126.0    32470.2 2577.0
Shell Scripts (1 concurrent)                   42.4    22965.7 5416.4
Shell Scripts (8 concurrent)                   6.0    3668.1 6113.5
System Call Overhead                         15000.0 1789273.3 1192.8
                                                               ========
System Benchmarks Index Score                                     3340.7

参考zhujiwiki得出，这分数比L5420x2强
然后加上intel吸牙膏的40%，完爆L5540
毕竟不可能不更新系统把，哈哈

下面是新买的1240的跑分，上半部分忘记复制了，记得是1316：

Dhrystone 2 using register variables    168945022.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone                   26219.9 MWIPS (10.0 s, 7 samples)
Execl Throughput                            26686.0 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       938942.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks       247646.8 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks    2897463.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                            5360592.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching             1373693.2 lps (10.0 s, 7 samples)
Process Creation                            72939.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                31378.1 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                4198.0 lpm (60.1 s, 2 samples)
System Call Overhead                      4343469.2 lps (10.0 s, 7 samples)

System Benchmarks Index Values             BASELINE    RESULT INDEX
Dhrystone 2 using register variables       116700.0  168945022.9  14476.9
Double-Precision Whetstone                      55.0    26219.9 4767.3
Execl Throughput                               43.0    26686.0 6206.1
File Copy 1024 bufsize 2000 maxblocks       3960.0    938942.5 2371.1
File Copy 256 bufsize 500 maxblocks          1655.0    247646.8 1496.4
File Copy 4096 bufsize 8000 maxblocks       5800.0 2897463.5 4995.6
Pipe Throughput                            12440.0 5360592.8 4309.2
Pipe-based Context Switching                4000.0 1373693.2 3434.2
Process Creation                               126.0    72939.5 5788.8
Shell Scripts (1 concurrent)                   42.4    31378.1 7400.5
Shell Scripts (8 concurrent)                   6.0    4198.0 6996.7
System Call Overhead                         15000.0 4343469.2 2895.6
                                                               ========
System Benchmarks Index Score                                     4631.8

作者: kyotrue 时间: 2018-1-18 10:21
独服还真用不着更新系统

作者: iamyg 时间: 2018-1-18 10:22

kyotrue 发表于 2018-1-18 10:21
独服还真用不着更新系统

为啥用不着

作者: iamyg 时间: 2018-1-18 10:31
手工置顶，大佬们快来

作者: 羽檬 时间: 2018-1-18 10:37
和e3比呢

作者: iamyg 时间: 2018-1-18 10:39

羽檬发表于 2018-1-18 10:37
和e3比呢

E3他家的不知道，估计跟v1是一个档次的吧
v2 3估计比不过

作者: haohaoge 时间: 2018-1-18 10:51
L5630 四千多具体忘了

作者: liusu 时间: 2018-1-18 10:54
提示: 作者被禁止或删除内容自动屏蔽

作者: 左手写爱 时间: 2018-1-18 11:04
大内存反正我是挺喜欢的

作者: kyotrue 时间: 2018-1-18 11:42

iamyg 发表于 2018-1-18 10:22
为啥用不着

不乱装软件就行了，整个服务器控制权在你自己手中啊。又不是小鸡，可能被其他小鸡读到你的内存

作者: icesky 时间: 2018-1-18 13:02
你要的intel
------------------------------------------------------------------------
Benchmark Run: Thu Jan 18 2018 10:56:56 - 11:25:14
16 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables    22903523.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone                   2780.0 MWIPS (9.9 s, 7 samples)
Execl Throughput                            1761.5 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       386941.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks       101641.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks    1090917.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                            532191.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching                82998.3 lps (10.0 s, 7 samples)
Process Creation                            3947.3 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                4772.1 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                2586.0 lpm (60.0 s, 2 samples)
System Call Overhead                      384742.8 lps (10.0 s, 7 samples)

System Benchmarks Index Values             BASELINE    RESULT INDEX
Dhrystone 2 using register variables       116700.0 22903523.9 1962.6
Double-Precision Whetstone                      55.0    2780.0 505.5
Execl Throughput                               43.0    1761.5 409.7
File Copy 1024 bufsize 2000 maxblocks       3960.0    386941.3 977.1
File Copy 256 bufsize 500 maxblocks          1655.0    101641.3 614.1
File Copy 4096 bufsize 8000 maxblocks       5800.0 1090917.9 1880.9
Pipe Throughput                            12440.0    532191.6 427.8
Pipe-based Context Switching                4000.0    82998.3 207.5
Process Creation                               126.0    3947.3 313.3
Shell Scripts (1 concurrent)                   42.4    4772.1 1125.5
Shell Scripts (8 concurrent)                   6.0    2586.0 4310.0
System Call Overhead                         15000.0    384742.8 256.5
                                                               ========
System Benchmarks Index Score                                        708.0

------------------------------------------------------------------------
Benchmark Run: Thu Jan 18 2018 11:25:14 - 11:53:14
16 CPUs in system; running 16 parallel copies of tests

Dhrystone 2 using register variables    191894521.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone                   36381.6 MWIPS (10.0 s, 7 samples)
Execl Throughput                            19008.3 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       468333.2 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks       133339.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks    1538501.1 KBps  (30.0 s, 2 samples)
Pipe Throughput                            5784169.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching             1318620.5 lps (10.0 s, 7 samples)
Process Creation                            42758.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                46531.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                6788.7 lpm (60.1 s, 2 samples)
System Call Overhead                      4356688.7 lps (10.0 s, 7 samples)

System Benchmarks Index Values             BASELINE    RESULT INDEX
Dhrystone 2 using register variables       116700.0  191894521.4  16443.4
Double-Precision Whetstone                      55.0    36381.6 6614.8
Execl Throughput                               43.0    19008.3 4420.5
File Copy 1024 bufsize 2000 maxblocks       3960.0    468333.2 1182.7
File Copy 256 bufsize 500 maxblocks          1655.0    133339.5 805.7
File Copy 4096 bufsize 8000 maxblocks       5800.0 1538501.1 2652.6
Pipe Throughput                            12440.0 5784169.7 4649.7
Pipe-based Context Switching                4000.0 1318620.5 3296.6
Process Creation                               126.0    42758.5 3393.5
Shell Scripts (1 concurrent)                   42.4    46531.9  10974.5
Shell Scripts (8 concurrent)                   6.0    6788.7  11314.6
System Call Overhead                         15000.0 4356688.7 2904.5
                                                               ========
System Benchmarks Index Score                                     4103.5

======= Script description and score comparison completed! =======

作者: iamyg 时间: 2018-1-18 13:09

icesky 发表于 2018-1-18 13:02
你要的intel
------------------------------------------------------------------------
Benchmark Run: ...

谢谢大佬，这是intel的哪个配置的
是5420还是5630还是1265的
能不能升级成最新的linux内核再跑一下
看看分数降不降
这里是个升级的教程：
https://www.duyaofei.com/2018/01 ... %90%AF%E7%94%A8bbr/

作者: icesky 时间: 2018-1-18 13:22

iamyg 发表于 2018-1-18 13:09
谢谢大佬，这是intel的哪个配置的
是5420还是5630还是1265的
能不能升级成最新的linux内核再跑一下

无聊去做这个，有好事者早就做了，结论是：“汇总出来的结论发上来。其实结论很简单，对于6、7、8三代影响不会很大（因为都是马甲，7500可以类比很多产品）。
从性能来说是会性能下滑肉眼可见，但是幅度很小，所以用1151四核的人可以不必太纠结。
按照目前得到的说法，六代酷睿之前的产品受影响比较大，越早的产品相对受影响会越大。给大家一个参考，4790K预计会有8%左右的性能下滑。”
另外，系统会逐渐优化的。

作者: iamyg 时间: 2018-1-18 13:39

icesky 发表于 2018-1-18 13:22
无聊去做这个，有好事者早就做了，结论是：“汇总出来的结论发上来。其实结论很简单，对于6、7、8三代影 ...

您的是啥配置的，5630吗？

作者: cdseoo 时间: 2018-1-18 14:45

iamyg 发表于 2018-1-18 13:09
谢谢大佬，这是intel的哪个配置的
是5420还是5630还是1265的
能不能升级成最新的linux内核再跑一下

闲的没事，跑了一下，昨天买的5630的结果：

============================================
Benchmark Run: Thu Jan 18 2018 13:24:32 - 13:52:51
16 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables    22784346.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone                   2763.5 MWIPS (10.0 s, 7 samples)
Execl Throughput                            1649.8 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       479817.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks       129251.1 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks    1259141.8 KBps  (30.0 s, 2 samples)
Pipe Throughput                            620379.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching                122768.1 lps (10.0 s, 7 samples)
Process Creation                            3642.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                3558.1 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                1780.7 lpm (60.0 s, 2 samples)
System Call Overhead                      482250.2 lps (10.0 s, 7 samples)

System Benchmarks Index Values             BASELINE    RESULT INDEX
Dhrystone 2 using register variables       116700.0 22784346.2 1952.4
Double-Precision Whetstone                      55.0    2763.5 502.5
Execl Throughput                               43.0    1649.8 383.7
File Copy 1024 bufsize 2000 maxblocks       3960.0    479817.8 1211.7
File Copy 256 bufsize 500 maxblocks          1655.0    129251.1 781.0
File Copy 4096 bufsize 8000 maxblocks       5800.0 1259141.8 2170.9
Pipe Throughput                            12440.0    620379.1 498.7
Pipe-based Context Switching                4000.0    122768.1 306.9
Process Creation                               126.0    3642.0 289.1
Shell Scripts (1 concurrent)                   42.4    3558.1 839.2
Shell Scripts (8 concurrent)                   6.0    1780.7 2967.8
System Call Overhead                         15000.0    482250.2 321.5
                                                               ========
System Benchmarks Index Score                                        741.0

------------------------------------------------------------------------
Benchmark Run: Thu Jan 18 2018 13:52:51 - 14:20:59
16 CPUs in system; running 16 parallel copies of tests

Dhrystone 2 using register variables    188898667.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone                   35899.7 MWIPS (10.0 s, 7 samples)
Execl Throughput                            22926.8 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       618169.1 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks       174694.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks    1874106.8 KBps  (30.0 s, 2 samples)
Pipe Throughput                            6966547.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching             1518616.0 lps (10.0 s, 7 samples)
Process Creation                            58798.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                36398.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                4883.1 lpm (60.1 s, 2 samples)
System Call Overhead                      4785320.7 lps (10.0 s, 7 samples)

System Benchmarks Index Values             BASELINE    RESULT INDEX
Dhrystone 2 using register variables       116700.0  188898667.1  16186.7
Double-Precision Whetstone                      55.0    35899.7 6527.2
Execl Throughput                               43.0    22926.8 5331.8
File Copy 1024 bufsize 2000 maxblocks       3960.0    618169.1 1561.0
File Copy 256 bufsize 500 maxblocks          1655.0    174694.0 1055.6
File Copy 4096 bufsize 8000 maxblocks       5800.0 1874106.8 3231.2
Pipe Throughput                            12440.0 6966547.3 5600.1
Pipe-based Context Switching                4000.0 1518616.0 3796.5
Process Creation                               126.0    58798.1 4666.5
Shell Scripts (1 concurrent)                   42.4    36398.8 8584.6
Shell Scripts (8 concurrent)                   6.0    4883.1 8138.4
System Call Overhead                         15000.0 4785320.7 3190.2
                                                               ========
System Benchmarks Index Score                                     4485.5

======= Script description and score comparison completed! =======

作者: iamyg 时间: 2018-1-18 15:16

cdseoo 发表于 2018-1-18 14:45
闲的没事，跑了一下，昨天买的5630的结果：

============================================

谢谢大佬
还是intel的划算，哈哈
大家都买intel的吧

作者: iamyg 时间: 2018-1-19 10:30
1240v1跑分如下，上面的忘记复制了，单核分数是1316.

Dhrystone 2 using register variables    168945022.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone                   26219.9 MWIPS (10.0 s, 7 samples)
Execl Throughput                            26686.0 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       938942.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks       247646.8 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks    2897463.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                            5360592.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching             1373693.2 lps (10.0 s, 7 samples)
Process Creation                            72939.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                31378.1 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                4198.0 lpm (60.1 s, 2 samples)
System Call Overhead                      4343469.2 lps (10.0 s, 7 samples)

System Benchmarks Index Values             BASELINE    RESULT INDEX
Dhrystone 2 using register variables       116700.0  168945022.9  14476.9
Double-Precision Whetstone                      55.0    26219.9 4767.3
Execl Throughput                               43.0    26686.0 6206.1
File Copy 1024 bufsize 2000 maxblocks       3960.0    938942.5 2371.1
File Copy 256 bufsize 500 maxblocks          1655.0    247646.8 1496.4
File Copy 4096 bufsize 8000 maxblocks       5800.0 2897463.5 4995.6
Pipe Throughput                            12440.0 5360592.8 4309.2
Pipe-based Context Switching                4000.0 1373693.2 3434.2
Process Creation                               126.0    72939.5 5788.8
Shell Scripts (1 concurrent)                   42.4    31378.1 7400.5
Shell Scripts (8 concurrent)                   6.0    4198.0 6996.7
System Call Overhead                         15000.0 4343469.2 2895.6
                                                               ========
System Benchmarks Index Score                                     4631.8

作者: tiane12 时间: 2018-1-19 11:05
看来都差不多啊，也不是AMD垃圾。

作者: iamyg 时间: 2018-1-19 11:07

tiane12 发表于 2018-1-19 11:05
看来都差不多啊，也不是AMD垃圾。

AMD还是比他们低20%左右呢

欢迎光临全球主机交流论坛 (https://hostloc.gdisk.cf/)