Unixbench Score on Core i5 750 with OpenVZ Kernel vs Xen/Openvz
This was unbelievable how much the Xen kernel slows things down, keep in mind both tests were done on the hostnode, one was with the Openvz-Xen hybrid kernel and the other was just OpenVZ. You can see the performance difference is nearly 300% better when not using the Xen kernel.
OpenVZ-Xen Kernel Test Results (I was wondering what was wrong/so slow with my Core i5!)
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
4 1 Based on the Byte Magazine Unix Benchmark
44 11
v v 4 4 1
v v 44444 1 v4.1 revisions mostly by David C. Niemi,
v 4 o 111 - WHT.2 Reston, VA, USA <niemi@tux.org>
WHT Variant by Andy A. Lee <andylee@dslextreme.com>
See: http://www.webhostingtalk.com/showthread.php?s=&threadid=308055
Dhrystone 2 using register variables 1 2
Double-Precision Whetstone 1 2
Execl Throughput 1
Filesystem Throughput 1024 bufsize 2000 maxblocks 1
Filesystem Throughput 256 bufsize 500 maxblocks 1
Filesystem Throughput 4096 bufsize 8000 maxblocks 1
Pipe Throughput 1 2
Pipe-based Context Switching 1 2
Process Creation 1
System Call Overhead 1 2
Shell Scripts (8 concurrent) 1
==============================================================
BYTE UNIX Benchmarks (Version 4.1-wht.2)
System -- Linux localhost.localdomain 2.6.18-194.3.1.el5.028stab069.5xen #1 SMP Tue May 18 18:33:35 MSD 2010 i686 i686 i386 GNU/Linux
/dev/sda3 19236340 177016 18082172 1% /mnt/sda3
Start Benchmark Run: Sat May 29 01:28:00 EDT 2010
01:28:00 up 2:38, 2 users, load average: 0.32, 0.10, 0.02
End Benchmark Run: Sat May 29 01:38:35 EDT 2010
01:38:35 up 2:48, 2 users, load average: 12.47, 5.83, 2.66
INDEX VALUES
TEST BASELINE RESULT INDEX
Dhrystone 2 using register variables 376783.7 24942262.7 662.0
Double-Precision Whetstone 83.1 1518.5 182.7
Execl Throughput 188.3 4822.1 256.1
File Copy 1024 bufsize 2000 maxblocks 2672.0 69636.0 260.6
File Copy 256 bufsize 500 maxblocks 1077.0 25286.0 234.8
File Read 4096 bufsize 8000 maxblocks 15382.0 829674.0 539.4
Pipe-based Context Switching 15448.6 149570.1 96.8
Pipe Throughput 111814.6 416802.5 37.3
Process Creation 569.3 7657.1 134.5
Shell Scripts (8 concurrent) 44.8 1084.6 242.1
System Call Overhead 114433.5 3620305.9 316.4
=========
FINAL SCORE 211.6
OpenVZ only Kernel,now the Core i5 shows its performance!
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
4 1 Based on the Byte Magazine Unix Benchmark
44 11
v v 4 4 1
v v 44444 1 v4.1 revisions mostly by David C. Niemi,
v 4 o 111 - WHT.2 Reston, VA, USA <niemi@tux.org>
WHT Variant by Andy A. Lee <andylee@dslextreme.com>
See: http://www.webhostingtalk.com/showthread.php?s=&threadid=308055
Dhrystone 2 using register variables 1 2
Double-Precision Whetstone 1 2
Execl Throughput 1
Filesystem Throughput 1024 bufsize 2000 maxblocks 1
Filesystem Throughput 256 bufsize 500 maxblocks 1
Filesystem Throughput 4096 bufsize 8000 maxblocks 1
Pipe Throughput 1 2
Pipe-based Context Switching 1 2
Process Creation 1
System Call Overhead 1 2
Shell Scripts (8 concurrent) 1
==============================================================
BYTE UNIX Benchmarks (Version 4.1-wht.2)
System -- Linux localhost.localdomain 2.6.18-194.3.1.el5.028stab069.5PAE #1 SMP Tue May 18 18:26:13 MSD 2010 i686 i686 i386 GNU/Linux
/dev/md0 297581760 1276412 280945208 1% /
Start Benchmark Run: Sat May 29 01:55:39 EDT 2010
01:55:39 up 2 min, 1 user, load average: 0.57, 0.30, 0.11
End Benchmark Run: Sat May 29 02:06:06 EDT 2010
02:06:06 up 13 min, 2 users, load average: 15.13, 6.61, 2.95
INDEX VALUES
TEST BASELINE RESULT INDEX
Dhrystone 2 using register variables 376783.7 24876280.4 660.2
Double-Precision Whetstone 83.1 1515.2 182.3
Execl Throughput 188.3 17397.6 923.9
File Copy 1024 bufsize 2000 maxblocks 2672.0 379937.0 1421.9
File Copy 256 bufsize 500 maxblocks 1077.0 107354.0 996.8
File Read 4096 bufsize 8000 maxblocks 15382.0 2918897.0 1897.6
Pipe-based Context Switching 15448.6 973442.9 630.1
Pipe Throughput 111814.6 3216279.5 287.6
Process Creation 569.3 50262.7 882.9
Shell Scripts (8 concurrent) 44.8 3636.8 811.8
System Call Overhead 114433.5 3680141.7 321.6
=========
FINAL SCORE 671.4
The difference is easy to see
Xen Kernel=211.6
OpenVZ Kernel=671.4
OpenVZ is more than 3x faster. I would hate to see the poor results inside a Xen container. How can Xen claim such high performance when this is clearly not the case?
/proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 30
model name : Intel(R) Core(TM) i5 CPU 750 @ 2.67GHz
stepping : 5
cpu MHz : 2666.780
cache size : 8192 KB
physical id : 0
siblings : 4
core id : 0
cpu cores : 4
apicid : 0
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 11
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx rdtscp lm constant_tsc ida nonstop_tsc pni monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr sse4_1 sse4_2 popcnt lahf_lm [8]
bogomips : 5333.56
processor : 1
vendor_id : GenuineIntel
cpu family : 6
model : 30
model name : Intel(R) Core(TM) i5 CPU 750 @ 2.67GHz
stepping : 5
cpu MHz : 2666.780
cache size : 8192 KB
physical id : 0
siblings : 4
core id : 1
cpu cores : 4
apicid : 2
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 11
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx rdtscp lm constant_tsc ida nonstop_tsc pni monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr sse4_1 sse4_2 popcnt lahf_lm [8]
bogomips : 5333.21
processor : 2
vendor_id : GenuineIntel
cpu family : 6
model : 30
model name : Intel(R) Core(TM) i5 CPU 750 @ 2.67GHz
stepping : 5
cpu MHz : 2666.780
cache size : 8192 KB
physical id : 0
siblings : 4
core id : 2
cpu cores : 4
apicid : 4
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 11
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx rdtscp lm constant_tsc ida nonstop_tsc pni monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr sse4_1 sse4_2 popcnt lahf_lm [8]
bogomips : 5333.24
processor : 3
vendor_id : GenuineIntel
cpu family : 6
model : 30
model name : Intel(R) Core(TM) i5 CPU 750 @ 2.67GHz
stepping : 5
cpu MHz : 2666.780
cache size : 8192 KB
physical id : 0
siblings : 4
core id : 3
cpu cores : 4
apicid : 6
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 11
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx rdtscp lm constant_tsc ida nonstop_tsc pni monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr sse4_1 sse4_2 popcnt lahf_lm [8]
bogomips : 5333.24
This system has a GA-H57M-USB3 motherboard, 4Gigs of DDR3 and a RAID1 Western Digital 2TB 64Meg cache disk array, running Centos 5.5 32 bit.
Tags:
unixbench, openvz, kernel, vs, xen, openvzthis, slows, hostnode, hybrid, byte, unix, benchmark, revisions, niemi, wht, reston, va, usa, tux, org, variant, andylee, dslextreme, http, www, webhostingtalk, showthread, php, threadid, dhrystone, register, variables, precision, whetstone, execl, throughput, filesystem, bufsize, maxblocks, context, switching, creation, overhead, shell, scripts, concurrent, benchmarks, linux, localhost, localdomain, smp, tue, msd, gnu, dev, sda, mnt, edt, users, index, baseline, pae, md, min, user, container, proc, cpuinfo, processor, vendor_id, genuineintel, cpu, intel, tm, ghz, stepping, mhz, cache, kb, siblings, cores, apicid, fdiv_bug, hlt_bug, f_bug, coma_bug, fpu, fpu_exception, cpuid, wp, flags, vme, pse, tsc, msr, mce, cx, apic, sep, mtrr, pge, mca, cmov, clflush, dts, acpi, mmx, fxsr, sse, ss, ht, pbe, nx, rdtscp, lm, constant_tsc, ida, nonstop_tsc, pni, ds_cpl, vmx, smx, est, ssse, xtpr, _, popcnt, lahf_lm, bogomips, ga, usb, motherboard, gigs, ddr, raid, western, digital, tb, meg, disk, array, centos,