Re: FreeBSD 9.1 vs CentOS 6.3 [WAS Re: freebsd-performance Digest, Vol 119, Issue 8]

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: FreeBSD 9.1 vs CentOS 6.3 [WAS Re: freebsd-performance Digest, Vol 119, Issue 8]

Davide D'Amico
Thank you Daniel for your tests, here my tests using sysbench v0.5 MySQL
Benchmarks r/w (80%/20%) test on 10.000.000 rows 2.000.000 query using
Standard OLTP: values represent the number of transactions per second
and the first number is obtained using 1 thread, the second one using 2
threads, 4, 8, 12, 16, 24, 32, 48 and 64 threads.

CentOS 6 5.6.10-ent:
4163 7653 10905 12511 13556 14832 16270 16733 16925 16895

VM CentOS 6 5.6.10-ent VMWare 5.1:
3201 5543 8299 12823 14331 15658 16842 15946 11529 9457

VM FreeBSD 9.1 5.6.10-ent VMWare 5.1 (*):
2102 3572 5917 8060 7905 7734 7104 7304 7612 7058

VM FreeBSD 9.1 5.6.10-ent VMWare 5.1 (**):
2026 3290 4927 ... (I stopped the tests because it seems similar to the
previous one)

FreeBSD 9.1 5.6.10-ent ZFS+SSD:
2780 4371 6876 8202 8077 7780 7563 7632 7960 8062

FreeBSD 9.1 5.6.10-ent ZFS tweaked+SSD:
2589 4679 6438 7073 7121 7227 7132 7273 7623 7672

Well, CentOS outperforms FreeBSD in every thread concurrency, and not
only in standard oltp tests.
I think I'll use CentOS for mysql servers.

Thank you for all your time spent, support and tests.

d.


(*)
Using:
   - sysctl.conf:
     - kern.eventtimer.periodic=1;
     - kern.timecounter.hardware=ACPI-fast;
   - loader.conf:
     - kern.hz=100;

(**)
Using:
   - sysctl.conf:
     - kern.eventtimer.periodic=1;
     - kern.timecounter.hardware=ACPI-fast;
   - loader.conf:
     - kern.hz=100;
   - malloc.conf -> 3N
_______________________________________________
[hidden email] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "[hidden email]"
Reply | Threaded
Open this post in threaded view
|

Re: FreeBSD 9.1 vs CentOS 6.3 [WAS Re: freebsd-performance Digest, Vol 119, Issue 8]

Davide D'Amico
Il 25/03/13 15:00, Davide D'Amico ha scritto:

> Thank you Daniel for your tests, here my tests using sysbench v0.5 MySQL
> Benchmarks r/w (80%/20%) test on 10.000.000 rows 2.000.000 query using
> Standard OLTP: values represent the number of transactions per second
> and the first number is obtained using 1 thread, the second one using 2
> threads, 4, 8, 12, 16, 24, 32, 48 and 64 threads.
>
> CentOS 6 5.6.10-ent:
> 4163 7653 10905 12511 13556 14832 16270 16733 16925 16895
>
> VM CentOS 6 5.6.10-ent VMWare 5.1:
> 3201 5543 8299 12823 14331 15658 16842 15946 11529 9457
>
> VM FreeBSD 9.1 5.6.10-ent VMWare 5.1 (*):
> 2102 3572 5917 8060 7905 7734 7104 7304 7612 7058
>
> VM FreeBSD 9.1 5.6.10-ent VMWare 5.1 (**):
> 2026 3290 4927 ... (I stopped the tests because it seems similar to the
> previous one)
>
> FreeBSD 9.1 5.6.10-ent ZFS+SSD:
> 2780 4371 6876 8202 8077 7780 7563 7632 7960 8062
>
> FreeBSD 9.1 5.6.10-ent ZFS tweaked+SSD:
> 2589 4679 6438 7073 7121 7227 7132 7273 7623 7672
>
> Well, CentOS outperforms FreeBSD in every thread concurrency, and not
> only in standard oltp tests.
> I think I'll use CentOS for mysql servers.
>
> Thank you for all your time spent, support and tests.
>
> d.
>
>
> (*)
> Using:
>    - sysctl.conf:
>      - kern.eventtimer.periodic=1;
>      - kern.timecounter.hardware=ACPI-fast;
>    - loader.conf:
>      - kern.hz=100;
>
> (**)
> Using:
>    - sysctl.conf:
>      - kern.eventtimer.periodic=1;
>      - kern.timecounter.hardware=ACPI-fast;
>    - loader.conf:
>      - kern.hz=100;
>    - malloc.conf -> 3N

Well, because of a misunderstanding the previous tests were related to
oltp.lua dataset/workload, using the oltp_simple I have:

VM FreeBSD 9.1 5.6.10-ent VMWare 5.1:
2919 4758 8661 14075 16436 16328 17172 17636 17926 18218

CentOS 6:
5677 11253 22129 32096 45800 47091 42608 13097 12979 13282

FreeBSD 9.1:
2874 5179 9154 13199 14291 11627 19766 19887 21197 21787

I don't know is these tests could help finding where the problem is, I
hope so.

I can do other test until wednesday 27/03 if you need.

Thanks,
d.

_______________________________________________
[hidden email] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "[hidden email]"
Reply | Threaded
Open this post in threaded view
|

Re: FreeBSD 9.1 vs CentOS 6.3 [WAS Re: freebsd-performance Digest, Vol 119, Issue 8]

Adrian Chadd-2
Can you please run a Linux install in a FreeBSD jail so we can see
whether it's the kernel or userland?

Thanks,




Adrian


On 25 March 2013 07:45, Davide D'Amico <[hidden email]> wrote:

> Il 25/03/13 15:00, Davide D'Amico ha scritto:
>
>> Thank you Daniel for your tests, here my tests using sysbench v0.5 MySQL
>> Benchmarks r/w (80%/20%) test on 10.000.000 rows 2.000.000 query using
>> Standard OLTP: values represent the number of transactions per second
>> and the first number is obtained using 1 thread, the second one using 2
>> threads, 4, 8, 12, 16, 24, 32, 48 and 64 threads.
>>
>> CentOS 6 5.6.10-ent:
>> 4163 7653 10905 12511 13556 14832 16270 16733 16925 16895
>>
>> VM CentOS 6 5.6.10-ent VMWare 5.1:
>> 3201 5543 8299 12823 14331 15658 16842 15946 11529 9457
>>
>> VM FreeBSD 9.1 5.6.10-ent VMWare 5.1 (*):
>> 2102 3572 5917 8060 7905 7734 7104 7304 7612 7058
>>
>> VM FreeBSD 9.1 5.6.10-ent VMWare 5.1 (**):
>> 2026 3290 4927 ... (I stopped the tests because it seems similar to the
>> previous one)
>>
>> FreeBSD 9.1 5.6.10-ent ZFS+SSD:
>> 2780 4371 6876 8202 8077 7780 7563 7632 7960 8062
>>
>> FreeBSD 9.1 5.6.10-ent ZFS tweaked+SSD:
>> 2589 4679 6438 7073 7121 7227 7132 7273 7623 7672
>>
>> Well, CentOS outperforms FreeBSD in every thread concurrency, and not
>> only in standard oltp tests.
>> I think I'll use CentOS for mysql servers.
>>
>> Thank you for all your time spent, support and tests.
>>
>> d.
>>
>>
>> (*)
>> Using:
>>    - sysctl.conf:
>>      - kern.eventtimer.periodic=1;
>>      - kern.timecounter.hardware=ACPI-fast;
>>    - loader.conf:
>>      - kern.hz=100;
>>
>> (**)
>> Using:
>>    - sysctl.conf:
>>      - kern.eventtimer.periodic=1;
>>      - kern.timecounter.hardware=ACPI-fast;
>>    - loader.conf:
>>      - kern.hz=100;
>>    - malloc.conf -> 3N
>
>
> Well, because of a misunderstanding the previous tests were related to
> oltp.lua dataset/workload, using the oltp_simple I have:
>
> VM FreeBSD 9.1 5.6.10-ent VMWare 5.1:
> 2919 4758 8661 14075 16436 16328 17172 17636 17926 18218
>
> CentOS 6:
> 5677 11253 22129 32096 45800 47091 42608 13097 12979 13282
>
> FreeBSD 9.1:
> 2874 5179 9154 13199 14291 11627 19766 19887 21197 21787
>
> I don't know is these tests could help finding where the problem is, I hope
> so.
>
> I can do other test until wednesday 27/03 if you need.
>
> Thanks,
> d.
>
>
> _______________________________________________
> [hidden email] mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-performance
> To unsubscribe, send any mail to
> "[hidden email]"
_______________________________________________
[hidden email] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "[hidden email]"
Reply | Threaded
Open this post in threaded view
|

Re: FreeBSD 9.1 vs CentOS 6.3 [WAS Re: freebsd-performance Digest, Vol 119, Issue 8]

Davide D'Amico
Il 25/03/13 18:11, Adrian Chadd ha scritto:
> Can you please run a Linux install in a FreeBSD jail so we can see
> whether it's the kernel or userland?

Sure, do you have a link on how to install gnu/linux on a fbsd jail?

Is ok if I use the VM I created in vmware (so it will be VMWARE ->
FreeBSD -> Linux Jail)?

d.

_______________________________________________
[hidden email] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "[hidden email]"