Benchmark for KVM Under AltixXE210

From Leo's Notes
Last edited on 29 March 2015, at 04:19.

Results Overview

Hardware AltixXE210
CPU 4x Intel(R) Xeon(R) CPU X5355 @ 2.66GHz
Hypervisor QEMU-KVM
Memory ?
Disk Soft RAID1 (mdadm) 2x 500GB SATA 2 7200RPM
Operating System ScientificLinux 6.3
Score 496.1 / 915.5

Raw Output

========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: localhost.localdomain: GNU/Linux
   OS: GNU/Linux -- 2.6.32-358.el6.x86_64 -- #1 SMP Fri Feb 22 00:31:26 UTC 2013
   Machine: x86_64 (x86_64)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: QEMU Virtual CPU version (cpu64-rhel6) (5320.0 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
   CPU 1: QEMU Virtual CPU version (cpu64-rhel6) (5320.0 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
   CPU 2: QEMU Virtual CPU version (cpu64-rhel6) (5320.0 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
   CPU 3: QEMU Virtual CPU version (cpu64-rhel6) (5320.0 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET
   01:44:08 up 4 min,  1 user,  load average: 0.42, 0.39, 0.17; runlevel 3

------------------------------------------------------------------------
Benchmark Run: Wed May 08 2013 01:44:08 - 02:12:14
4 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       23243980.0 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     1512.5 MWIPS (10.0 s, 7 samples)
Execl Throughput                                488.4 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        499350.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          151360.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        791721.3 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1028436.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  94362.3 lps   (10.0 s, 7 samples)
Process Creation                               1055.6 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   1323.3 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    333.8 lpm   (60.1 s, 2 samples)
System Call Overhead                        1201269.0 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   23243980.0   1991.8
Double-Precision Whetstone                       55.0       1512.5    275.0
Execl Throughput                                 43.0        488.4    113.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     499350.0   1261.0
File Copy 256 bufsize 500 maxblocks            1655.0     151360.3    914.6
File Copy 4096 bufsize 8000 maxblocks          5800.0     791721.3   1365.0
Pipe Throughput                               12440.0    1028436.1    826.7
Pipe-based Context Switching                   4000.0      94362.3    235.9
Process Creation                                126.0       1055.6     83.8
Shell Scripts (1 concurrent)                     42.4       1323.3    312.1
Shell Scripts (8 concurrent)                      6.0        333.8    556.4
System Call Overhead                          15000.0    1201269.0    800.8
                                                                   ========
System Benchmarks Index Score                                         496.1

------------------------------------------------------------------------
Benchmark Run: Wed May 08 2013 02:12:14 - 02:40:26
4 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables       91942956.4 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     6045.8 MWIPS (10.0 s, 7 samples)
Execl Throughput                               1506.0 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        277706.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           83388.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        661509.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             4101639.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 302829.6 lps   (10.0 s, 7 samples)
Process Creation                               2485.2 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2604.9 lpm   (60.1 s, 2 samples)
Shell Scripts (8 concurrent)                    336.8 lpm   (60.6 s, 2 samples)
System Call Overhead                        2503841.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   91942956.4   7878.6
Double-Precision Whetstone                       55.0       6045.8   1099.2
Execl Throughput                                 43.0       1506.0    350.2
File Copy 1024 bufsize 2000 maxblocks          3960.0     277706.3    701.3
File Copy 256 bufsize 500 maxblocks            1655.0      83388.4    503.9
File Copy 4096 bufsize 8000 maxblocks          5800.0     661509.0   1140.5
Pipe Throughput                               12440.0    4101639.7   3297.1
Pipe-based Context Switching                   4000.0     302829.6    757.1
Process Creation                                126.0       2485.2    197.2
Shell Scripts (1 concurrent)                     42.4       2604.9    614.4
Shell Scripts (8 concurrent)                      6.0        336.8    561.4
System Call Overhead                          15000.0    2503841.1   1669.2
                                                                   ========
System Benchmarks Index Score                                         915.5

Other Information

This should be comparable to the virtual machine under ESXi as it was given the same set of resources. It would appear that VMWare has slightly faster CPU performance.

See Also