Solved

Benchmarking cloud hosting

Posted on 2014-02-22
13
366 Views
Last Modified: 2014-03-10
Hi,

I want to ask how do  know if I am getting desired resources commited by webhosting provider,

How do I know if I am getting cpu core with full access without any priority assignment or if I am sharing cpu cores on my cloud?

I have asked dediserve if they assign any priority / share limits, they claim to give full access to cores, however I need to authenticate and verify their claims.

the output for cat /proc/cpuinfo {But I am not satisfied with the server performance }, would appreciate help


processor      : 0
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 1
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 2
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 3
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 4
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 5
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 6
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 7
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 8
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 9
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 10
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 11
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 12
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 13
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 14
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 15
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:

processor      : 16
vendor_id      : GenuineIntel
cpu family      : 6
model            : 45
model name      : Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz
stepping      : 7
cpu MHz            : 2200.050
cache size      : 20480 KB
fpu            : yes
fpu_exception      : yes
cpuid level      : 13
wp            : yes
flags            : fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm rep_good aperfmperf unfair_spinlock pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes hypervisor lahf_lm ida arat epb xsaveopt pln pts dts
bogomips      : 4400.10
clflush size      : 64
cache_alignment      : 64
address sizes      : 46 bits physical, 48 bits virtual
power management:
0
Comment
Question by:anamarif
  • 5
  • 5
  • 2
  • +1
13 Comments
 
LVL 13

Expert Comment

by:Sandy
ID: 39880266
this is just processor listing.. to check the exact utilization configure any small app and try to inject from outside.

for benchmark testing you can also refer below link.

https://wiki.archlinux.org/index.php/benchmarking

TY/SA
0
 

Author Comment

by:anamarif
ID: 39880280
sandy, what I want to know is if cpu cores are operating at their rated GHZ and are all functioning as they should, the server is already in production, having a wordpress blog with some visitors as well.
However, the cpu load seems that server is not performing as it should.

I don't know if unixbench is something good to go.
0
 
LVL 118
ID: 39880332
Are you hosted on a Dedicaterd Server, or are you operating under a hypervisor?

The issue you have is most hypervisor pass the instruction directly to the Host CPU, e.g. the CPU is not emulated, your instructions are being executed directly on a physical Core of the Host CPU.

and it would be difficult to detect, unless you run the same benchmark, against a physical processor.

One possibility, can you check for the precense of Intel VT?
0
 

Author Comment

by:anamarif
ID: 39880422
how do I check presence of intel vT?

My question is particularly about dediserve.com and it's cloud hosting solution, most probably hypervisor.
0
 
LVL 61

Expert Comment

by:gheist
ID: 39880449
Actually you run under a hypervisor.
Since you mention wordpress - it is single-threaded and very latency-sensitive. Thowing 2x more 2x faster CPUs at it will not make it any faster.
FYI - cloud means virtualisation.

Do you need help with tuning wordpress or you will google it out yourself?
Fairly responsive wordpress site can run on free amazon instance.... (i.e 600MBRRAM + sub-standard CPU & disk)
0
 

Author Comment

by:anamarif
ID: 39880452
wordpress site may not be faster but the server load shouldn't be higher as well? also mysql performance shouldnt be degraded as the traffic increases or decreases, right?
0
Zoho SalesIQ

Hassle-free live chat software re-imagined for business growth. 2 users, always free.

 
LVL 61

Expert Comment

by:gheist
ID: 39880489
Wordpress use is not exactly "no load" for server.
Lets tune from ground up:
how many IO/s your disk does
what is the throughput?
?-> what is the optimal blocksize in the end?

mysql -> mysqltuner, tuning-primer (at least one shouls be with system packages)
make sure it does not deviate too much from optimal blocksize when accessing the disk
large-pages !!! you are under virtualisation, you get much better memory performance from them.

apache+php
If you run apache prefork and php in-process each parallel connection keepd DB connection open.
If you switch to apache-worker or nginx and php via fcgid that connection number falls 10x

Now wordpress - toss the coin which of super-duper optimizer plugin you want...

On hosting: consider you have 10-20 VMs of similar size in same computer, normally you get at least fair share of CPU time at any given moment, it sums up quite well for all.
0
 
LVL 13

Expert Comment

by:Sandy
ID: 39880496
Their are several tools which can be used.

TurboStat
cpufreq
cpupowerutils

TY/SA
0
 
LVL 61

Expert Comment

by:gheist
ID: 39880520
That does not bypass virtualisation....
One has to run some CPU benchmark continuously to get the idea of CPU resource part they can get.
0
 

Author Comment

by:anamarif
ID: 39880710
@gheist, thank you for your feedback, I highly appreciate it, you seems quite good in all.
could you please tell me how do I monitor/test/benchmark IO, THROUHPUT,OPTIMAL BLOCKSIZE.
my question basically surrounds about benchmarking resources like cpu/ram/hdd like performance test, unix bench etc....
0
 
LVL 61

Accepted Solution

by:
gheist earned 500 total points
ID: 39881143
time dd conv=sync bs=4k of=/home/bench count=1024 ; rm -f /home/bench
divide 4G by time spent and get MB/s
divide 4G by IO count get IO/s

Now double blocksize and halve count
...

Now paste that in some table processor....
MB/s graph will grow until bandwidth is reached
OR IO/s will be steady until it starts to lower when it reaches disk bandwidth
Once you have IO/s and MB/s you can make "cleaner" graph and say 128K blocks are best for DB IO (it really depends, sweet spot can be anywhere between 4K (local disk) and 4M on high latency NAS)

For common disk it is 4K, maybe 8K for Advanced format disk, few stripes per RAID, varies for network storage (usually not less than 8K given size of jumbo frame, for high latency much more)
0
 
LVL 61

Expert Comment

by:gheist
ID: 39881149
Since you are on the webserver
Benchmark RAM using same method as disk (using tmpfs i.e mount -t tmpfs -o noatime /mnt /mnt)
Benchmark CPU using openssl (but since you have no reference...) as only high CPU use may be https:// on a web server
If you focus on wordpress it has some benchmark plugins - you may wish to compare with simlar class CPUs.
0
 

Author Closing Comment

by:anamarif
ID: 39919669
Thanks
0

Featured Post

Want to promote your upcoming event?

Attending an event? Speaking at a conference? Or exhibiting at a tradeshow? Easily inform your contacts by using a promotional banner in your email signature. This will ensure your organization’s most important contacts are in the know.

Join & Write a Comment

#Citrix #XenApp #Citrix Scout #Citrix Insight Services #Microsoft VMMAP #Microsoft ADEXPLORE #Microsoft RAMMAP #Microsoft TCPVIEW #Microsoft AUTORUNS #Microsoft PROCESS EXPLORER #Microsoft PROCESS MONITOR
Will try to explain how to use the VMware feature TAGs in the VMs and create Veeam Backup Jobs using TAGs. Since this article is too long, I will create second article for the Veeam tasks.
Connecting to an Amazon Linux EC2 Instance from Windows Using PuTTY.
This Micro Tutorial will explain how to export DynamoDB tables in Amazon Web Services.

746 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question

Need Help in Real-Time?

Connect with top rated Experts

11 Experts available now in Live!

Get 1:1 Help Now