Re: mysql performance test results under FreeBSD-7.0-CURRENT

From: Ganbold <ganbold_at_micom.mng.net>
Date: Wed, 12 Apr 2006 14:20:19 +0900
Hi,

I did couple of tests with mysql-4.1.18 (from ports).

gw#pkg_info|grep mysql
mysql-client-4.1.18_1 Multithreaded SQL database (client)
mysql-server-4.1.18_2 Multithreaded SQL database (server)

On today's CURRENT
(FreeBSD gw.micom.mng.net 7.0-CURRENT FreeBSD 7.0-CURRENT #20: Wed Apr 
12 11:59:54 ULAST 2006     
tsgan_at_gw.micom.mng.net:/usr/obj/usr/src/sys/GW  i386)
I got these numbers:

1. mysql(BUILD_OPTIMIZED=yes, BUILD_STATIC=yes, 
WITH_PROC_SCOPE_PTH=yes)-libpthread-tsc-sched_4bsd+preemption

select_index    20000   0       0       10096.35
select_index    20000   0       0       11575.97
select_index    20000   0       0       11715.69
select_index    20000   0       0       11299.58
select_index    20000   0       0       11538.93
select_index    10000   2       0       1681.09
select_index    10000   2       0       1808.25
select_index    10000   2       0       1798.90
select_index    10000   3       0       1801.85
select_index    10000   5       0       1811.03

2. mysql(BUILD_OPTIMIZED=yes, BUILD_STATIC=yes, 
WITH_PROC_SCOPE_PTH=yes)-libpthread-acpifast-sched_4bsd+preemption

select_index    20000   0       0       9440.26
select_index    20000   0       0       10454.90
select_index    20000   0       0       10586.03
select_index    20000   0       0       10488.21
select_index    20000   0       0       10577.91
select_index    10000   4       0       1724.35
select_index    10000   4       0       1685.37
select_index    10000   7       0       1688.11
select_index    10000   8       0       1715.86
select_index    10000   7       0       1591.31

Ganbold

Michael Vince wrote:
> Ganbold wrote:
>> Hi all,
>>
>> I did make some mysql performance tests under FreeBSD-7.0-CURRENT 
>> with various
>> scheduler and compile time options.
>> It seems like mysql(BUILD_OPTIMIZED=yes, BUILD_STATIC=yes, 
>> WITH_PROC_SCOPE_PTH=yes)-libpthread-tsc-sched_4bsd+preemption gives
>> better performance.
>>
>> The test results are at:
>>
>> http://www.mnbsd.org/ftp/mysql_test_results.txt
>>
>> There are several things I didn't test and this leads to some questions:
>>
>> 1. I didn't make test with Poul-Henning's CPU accounting patch. 
>> Somehow I can't apply it 
>> (http://phk.freebsd.dk/patch/cpu_acct_2.patch) cleanly. Where can I 
>> find latest patch?
>> When this patch will be included in CURRENT?
>>
>> 2. I didn't make test with Robert Watson's patch 
>> (http://www.watson.org/~robert/freebsd/clock/)? Does CURRENT src tree 
>> include it? If not when this patch will be included in CURRENT?
>>
>> 3. I did make tests with default malloc in CURRENT. I'm confused what 
>> malloc options should try (jemalloc? phkmalloc?) What is the default 
>> malloc in CURRENT? How to use these different mallocs?
>>
>> thanks in advance,
>> Ganbold
>
> Nice work!
>
> I think it would be equally interesting if you posted benchmarks for 
> MySQL 4.1.x
> Since the last post comparing MySQL 5.x vs 4.1.x on FreeBSD showed 
> that MySQL 4.1 series had almost double the benchmark performance 
> numbers over 5.x MySQL
> I couldn't bare the idea of using MySQL 5 if its still only half as 
> fast and has features I don't need.
> I believe most people would prefer 4.1 unless they really need the 
> features that are in 5.x or don't need performance at all.
>
> Mike
>
>
>
>
>
> _______________________________________________
> freebsd-current_at_freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to 
> "freebsd-current-unsubscribe_at_freebsd.org"
>
>
>
Received on Wed Apr 12 2006 - 03:20:53 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:54 UTC