全球主机交流论坛
标题:
嵌套虚拟机CPU性能测试跑分
[打印本页]
作者:
2019年
时间:
2023-11-7 14:18
标题:
嵌套虚拟机CPU性能测试跑分
本帖最后由 2019年 于 2023-11-7 15:50 编辑
因为同时需要hyper-v pve
没办法,再hyper-v里又套了个pve
这个跑分大概损失了多少性能? 7950x3d 2次虚拟后的8核心小鸡
========================================================================
unixbench
------------------------------------------------------------------------
Benchmark Run: Tue Nov 07 2023 14:33:40 - 15:01:42
8 CPUs in system; running 1 parallel copy of tests 单线程测试
Dhrystone 2 using register variables 13723963.7 lps (10.0 s, 7 samples)
Double-Precision Whetstone 1534.2 MWIPS (9.1 s, 7 samples)
Execl Throughput 456.6 lps (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 321263.8 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 84388.7 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 930805.4 KBps (30.0 s, 2 samples)
Pipe Throughput 526047.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching 11076.7 lps (10.0 s, 7 samples)
Process Creation 1976.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 835.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 537.7 lpm (60.1 s, 2 samples)
System Call Overhead 502625.8 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 13723963.7 1176.0
Double-Precision Whetstone 55.0 1534.2 278.9
Execl Throughput 43.0 456.6 106.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 321263.8 811.3
File Copy 256 bufsize 500 maxblocks 1655.0 84388.7 509.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 930805.4 1604.8
Pipe Throughput 12440.0 526047.8 422.9
Pipe-based Context Switching 4000.0 11076.7 27.7
Process Creation 126.0 1976.1 156.8
Shell Scripts (1 concurrent) 42.4 835.5 197.0
Shell Scripts (8 concurrent) 6.0 537.7 896.1
System Call Overhead 15000.0 502625.8 335.1
========
System Benchmarks Index Score 341.5
------------------------------------------------------------------------
Benchmark Run: Tue Nov 07 2023 15:01:42 - 15:30:13
8 CPUs in system; running 8 parallel copies of tests 多线程测试
Dhrystone 2 using register variables 65572144.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 9601.4 MWIPS (9.9 s, 7 samples)
Execl Throughput 4122.1 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 300356.2 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 80606.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 712416.3 KBps (30.0 s, 2 samples)
Pipe Throughput 2301409.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching 384914.9 lps (10.0 s, 7 samples)
Process Creation 10919.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 7044.8 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 1033.2 lpm (60.3 s, 2 samples)
System Call Overhead 2029593.1 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 65572144.5 5618.9
Double-Precision Whetstone 55.0 9601.4 1745.7
Execl Throughput 43.0 4122.1 958.6
File Copy 1024 bufsize 2000 maxblocks 3960.0 300356.2 758.5
File Copy 256 bufsize 500 maxblocks 1655.0 80606.5 487.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 712416.3 1228.3
Pipe Throughput 12440.0 2301409.7 1850.0
Pipe-based Context Switching 4000.0 384914.9 962.3
Process Creation 126.0 10919.0 866.6
Shell Scripts (1 concurrent) 42.4 7044.8 1661.5
Shell Scripts (8 concurrent) 6.0 1033.2 1722.0
System Call Overhead 15000.0 2029593.1 1353.1
========
System Benchmarks Index Score 1309.7
======= Script description and score comparison completed! =======
----------------------------------------------------------------------
CPU Model : AMD Ryzen 9 7950X3D 16-Core Processor
CPU Cores : 4 @ 4192.022 MHz
CPU Cache : 1024 KB
AES-NI : ✓ Enabled
VM-x/AMD-V : ✗ Disabled
Total Disk : 1000.7 GB (9.6 GB Used)
Total Mem : 8.4 GB (1.8 GB Used)
Total Swap : 4.3 GB (0 Used)
System uptime : 2 days, 1 hour 35 min
Load average : 0.00, 0.01, 0.05
OS : CentOS Linux release 7.9.2009 (Core)
Arch : x86_64 (64 Bit)
Kernel : 3.10.0-1160.102.1.el7.x86_64
TCP CC : cubic
Virtualization : Microsoft Virtual Machine
IPv4/IPv6 : ✓ Online / ✓ Online
Organization : AS55799 IPTELECOM ASIA
Location : Los Angeles / US
Region : California
----------------------------------------------------------------------
I/O Speed(1st run) : 3.6 GB/s
I/O Speed(2nd run) : 3.4 GB/s
I/O Speed(3rd run) : 3.3 GB/s
I/O Speed(average) : 3515.7 MB/s
嵌套一层的io 3.6G
嵌套两层的io 600mb
IO损失很大,cpu跑分我看不懂,再网上找了一下其他u的跑分,估计损失80%性能
CPU Model : Common KVM processor
CPU Cores : 8 @ 4192.012 MHz
CPU Cache : 512 KB
AES-NI : ✗ Disabled
VM-x/AMD-V : ✗ Disabled
Total Disk : 305.5 GB (63.8 GB Used)
Total Mem : 20.4 GB (2.5 GB Used)
Total Swap : 15.4 GB (0 Used)
System uptime : 0 days, 2 hour 37 min
Load average : 0.09, 4.32, 7.25
OS : CentOS Linux release 7.9.2009 (Core)
Arch : x86_64 (64 Bit)
Kernel : 3.10.0-1160.el7.x86_64
TCP CC : cubic
Virtualization : KVM
IPv4/IPv6 : ✓ Online / ✗ Offline
Organization : AS55799 IPTELECOM ASIA
Location : Los Angeles / US
Region : California
----------------------------------------------------------------------
I/O Speed(1st run) : 532 MB/s
I/O Speed(2nd run) : 822 MB/s
I/O Speed(3rd run) : 641 MB/s
I/O Speed(average) : 665.0 MB/s
作者:
【菠萝云】
时间:
2023-11-7 15:01
小鸡里开小鸡，性能损失非常大
作者:
闻风听雨
时间:
2023-11-7 15:04
hyper-v开嵌套虚拟化了吗
作者:
yjx79802
时间:
2023-11-7 15:20
子子孙孙，无穷匮也。为什么不试试用docker跑linux呢，我之前用一个高性能VPS不支持虚拟化了，我就用的docker，docker里面可以安装openwrt,装debian,装ubuntu,都行啊。。
作者:
PA-fan
时间:
2023-11-7 15:21
win就用hyperv，linux就用pve，你嵌套用的理由是啥
作者:
zhouktv1
时间:
2023-11-7 15:22
修改pve里小鸡CPu配置，你看你的型号是common这个能测毛分数啊
Send by DZ Reader
作者:
2019年
时间:
2023-11-7 15:36
zhouktv1 发表于 2023-11-7 15:22
修改pve里小鸡CPu配置，你看你的型号是common这个能测毛分数啊
Send by DZ Reader
更新了 大佬帮看看
作者:
func
时间:
2023-11-7 16:04
欢迎光临 全球主机交流论坛 (https://loc.888543.xyz/)
Powered by Discuz! X3.4
