声明:本测试纯技术研究与分析,不附加任何其它目的,仅为本人笔记。
原创内容,任何转载请务必注明出处,否则必追究。
| 主机名: | svr01 |
| 备注: | - |
| 机型: | 小型主机( 4ECU / 2048内存 / 30GB 硬盘) |
| 镜像: | linux centos54 64 v111110 |
| 安全策略: | 22 |
| 物理地址: | 00:16:3e:01:05:b1 |
| 内网IP: | 172.17.12.xx |
| 内网掩码: | 255.255.192.0 |
| 外网IP: | 118.244.225.45 |
| 外网域名: | vm-172-17-12-xx.hb01.sdcloud.cn |
| 主机名: | svr02 |
| 备注: | - |
| 机型: | 小型主机( 4ECU / 2048内存 / 30GB 硬盘) |
| 镜像: | linux ubuntu1204 32 20121017 |
| 安全策略: | 22 |
| 物理地址: | 00:16:3e:01:06:6e |
| 内网IP: | 172.17.12.xx |
| 内网掩码: | 255.255.192.0 |
| 外网IP: | 211.101.12.222 |
| 外网域名: | vm-172-17-12-xx.hb01.sdcloud.cn |
| 主机名: | svr03 |
| 备注: | - |
| 机型: | 小型主机( 4ECU / 2048内存 / 30GB 硬盘) |
| 镜像: | linux centos62 64 |
| 安全策略: | 22 |
| 物理地址: | 00:16:3e:01:01:e9 |
| 内网IP: | 172.17.10.xx |
| 内网掩码: | 255.255.192.0 |
| 外网IP: | 211.101.1.228 |
| 外网域名: | vm-172-17-10-xx.hb01.sdcloud.cn |
问题: svr01、svr03在进行两项CPU测试时,cpu均无法打满,总是在低值波动。但在大素数测试中结果表现差异较大,考虑系统或生产环境的原因。
测试情况统计:
主机svr01 CPU性能测试(寻找大素数)
find prime during 2 - 20000000 : 时间过长放弃
主机svr01 CPU测试(UnixBench 5.1.3)
========================================================================
Benchmark Run: Fri Dec 28 2012 15:58:46 - 16:28:45
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 7179260.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3058.7 MWIPS (7.1 s, 7 samples)
Execl Throughput 760.5 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 107427.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 28072.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 322849.7 KBps (30.0 s, 2 samples)
Pipe Throughput 147751.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 43900.1 lps (10.0 s, 7 samples)
Process Creation 1522.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1468.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 203.1 lpm (60.1 s, 2 samples)
System Call Overhead 123872.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 7179260.5 615.2
Double-Precision Whetstone 55.0 3058.7 556.1
Execl Throughput 43.0 760.5 176.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 107427.5 271.3
File Copy 256 bufsize 500 maxblocks 1655.0 28072.4 169.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 322849.7 556.6
Pipe Throughput 12440.0 147751.9 118.8
Pipe-based Context Switching 4000.0 43900.1 109.8
Process Creation 126.0 1522.7 120.8
Shell Scripts (1 concurrent) 42.4 1468.5 346.4
Shell Scripts (8 concurrent) 6.0 203.1 338.5
System Call Overhead 15000.0 123872.5 82.6
========
System Benchmarks Index Score 231.5
主机svr01 IO测试(dd)
root@SNDA-172-17-12-58:/home# time dd if=/dev/zero of=/data/test.img bs=1k count=8192000
8388608000 bytes (8.4 GB) copied, 54.468 s, 154 MB/s
[root@SNDA-172-17-12-28 home]# time dd if=/dev/zero of=/data/test.img bs=8k count=1024000
8388608000 bytes (8.4 GB) copied, 46.1496 seconds, 182 MB/s
[root@SNDA-172-17-12-28 home]# time dd if=/dev/zero of=/data/test.img bs=256k count=32768
8589934592 bytes (8.6 GB) copied, 33.017 seconds, 260 MB/s
[root@SNDA-172-17-12-28 home]# time dd if=/dev/zero of=/data/test.img bs=1MB count=8192
8192000000 bytes (8.2 GB) copied, 29.4471 seconds, 278 MB/s
主机svr01 IO测试(iozone)
Command line used: ./iozone -r16k -P0 -s512m -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
524288 16 341030 719114 1747114 1591669 1120228 688255 1286915
Command line used: ./iozone -r16k -P0 -s1G -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
1048576 16 209891 652653 1679314 1471155 1073943 583257 1132575
Command line used: ./iozone -I -r16k -P0 -s512m -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
524288 16 106975 160305 122218 121797 7986 15207 9642
Command line used: ./iozone -I -r16k -P0 -s1G -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
1048576 16 113686 165463 87785 65743 3901 8404 19313
主机svr02 CPU性能测试(寻找大素数)
find prime during 2 - 20000000 using: 343684 ms
主机svr02 CPU性能测试(UnixBench 5.1.3)
========================================================================
Benchmark Run: Fri Dec 28 2012 15:01:07 - 15:29:42
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 8291582.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 1395.8 MWIPS (10.4 s, 7 samples)
Execl Throughput 1544.5 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 313859.6 KBps (30.1 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 86153.0 KBps (30.1 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 707569.5 KBps (30.0 s, 2 samples)
Pipe Throughput 428021.1 lps (10.5 s, 7 samples)
Pipe-based Context Switching 101968.1 lps (10.0 s, 7 samples)
Process Creation 6207.3 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 3128.2 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 409.3 lpm (60.1 s, 2 samples)
System Call Overhead 1541982.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 8291582.4 710.5
Double-Precision Whetstone 55.0 1395.8 253.8
Execl Throughput 43.0 1544.5 359.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 313859.6 792.6
File Copy 256 bufsize 500 maxblocks 1655.0 86153.0 520.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 707569.5 1219.9
Pipe Throughput 12440.0 428021.1 344.1
Pipe-based Context Switching 4000.0 101968.1 254.9
Process Creation 126.0 6207.3 492.6
Shell Scripts (1 concurrent) 42.4 3128.2 737.8
Shell Scripts (8 concurrent) 6.0 409.3 682.1
System Call Overhead 15000.0 1541982.7 1028.0
========
System Benchmarks Index Score 547.7
主机svr02 IO测试(dd)
root@SNDA-172-17-12-58:/home# time dd if=/dev/zero of=/data/test.img bs=1KB count=8192000
8192000000 bytes (8.2 GB) copied, 54.2757 s, 151 MB/s
root@SNDA-172-17-12-58:/home# time dd if=/dev/zero of=/data/test.img bs=8k count=1024000
8388608000 bytes (8.4 GB) copied, 40.9344 s, 205 MB/s
root@SNDA-172-17-12-58:/home# time dd if=/dev/zero of=/data/test.img bs=256k count=32768
8589934592 bytes (8.6 GB) copied, 39.4094 s, 218 MB/s
root@SNDA-172-17-12-58:/home# time dd if=/dev/zero of=/data/test.img bs=1MB count=8192
8192000000 bytes (8.2 GB) copied, 34.8569 s, 235 MB/s
主机svr02 IO测试(iozone )
Command line used: ./iozone -r16k -P0 -s512m -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
524288 16 234612 108695 2331495 2037993 1607081 32405 1952335
Command line used: ./iozone -r16k -P0 -s1G -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
1048576 16 185217 189748 2370621 2076812 1678763 11132 1878198
Command line used: ./iozone -I -r16k -P0 -s512m -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
524288 16 128107 91400 138879 143835 9291 12474 17084
Command line used: ./iozone -I -r16k -P0 -s1G -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
1048576 16 114410 119887 143978 85175 3638 7588 14241
主机svr03 CPU性能测试(寻找大素数)
find prime during 2 - 20000000 using: 167219 ms
主机svr03 CPU性能测试(UnixBench 5.1.3)
========================================================================
Benchmark Run: Fri Dec 28 2012 18:16:22 - 18:47:13
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 12107222.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2948.6 MWIPS (9.9 s, 7 samples)
Execl Throughput 588.5 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 94116.4 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 24749.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 305443.5 KBps (30.1 s, 2 samples)
Pipe Throughput 148183.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching 30945.4 lps (10.0 s, 7 samples)
Process Creation 1241.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1112.7 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 150.1 lpm (60.4 s, 2 samples)
System Call Overhead 145249.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 12107222.2 1037.5
Double-Precision Whetstone 55.0 2948.6 536.1
Execl Throughput 43.0 588.5 136.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 94116.4 237.7
File Copy 256 bufsize 500 maxblocks 1655.0 24749.4 149.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 305443.5 526.6
Pipe Throughput 12440.0 148183.3 119.1
Pipe-based Context Switching 4000.0 30945.4 77.4
Process Creation 126.0 1241.2 98.5
Shell Scripts (1 concurrent) 42.4 1112.7 262.4
Shell Scripts (8 concurrent) 6.0 150.1 250.1
System Call Overhead 15000.0 145249.5 96.8
========
System Benchmarks Index Score 212.0
主机svr03 IO测试(dd)
[root@SNDA-172-17-10-248 home]# time dd if=/dev/zero of=/data/test.img bs=1k count=8192000
8388608000 bytes (8.4 GB) copied, 127.819 s, 65.6 MB/s
[root@SNDA-172-17-10-248 home]# time dd if=/dev/zero of=/data/test.img bs=8k count=1024000
8388608000 bytes (8.4 GB) copied, 51.1438 s, 164 MB/s
[root@SNDA-172-17-10-248 home]# time dd if=/dev/zero of=/data/test.img bs=256k count=32768
8589934592 bytes (8.6 GB) copied, 42.0778 s, 204 MB/s
[root@SNDA-172-17-10-248 home]# time dd if=/dev/zero of=/data/test.img bs=1MB count=8192
8192000000 bytes (8.2 GB) copied, 37.5368 s, 218 MB/s
主机svr03 IO测试
Command line used: ./iozone -r16k -P0 -s512m -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
524288 16 191420 346583 1909348 1912503 1384704 213525 1379393
Command line used: ./iozone -r16k -P0 -s1G -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
1048576 16 127795 300081 1900796 1708446 1176290 15589 1454434
Command line used: ./iozone -I -r16k -P0 -s512m -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
524288 16 76516 136381 142755 126256 12236 13244 15596
Command line used: ./iozone -I -r16k -P0 -s1G -i0 -i1 -i2 -i5 -i8
random random bkwd record stride
KB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread
1048576 16 96240 138632 134615 143562 4683 8133 16112
1352

被折叠的 条评论
为什么被折叠?



