Re: MySQL performance problem on MacOS X

On 6/7/05, Adam Dennis <adam@xxxxxxxxxxxxx> wrote:

> This worried me so I did some internet research and the results are a
> little more reassuring:

Me too, but I am not that reassured...

> 1. Anandtech screwed up by using a crippled compiler for the G5 test
> and have misunderstandings of process / thread handling in OS X.
> (see comments for anandtech article and http://ridiculousfish.com/blog/?p=17)

I have the same symptoms with Tenon' MySQL, is it also badly compiled ?

> 2. Using Apache2 with it's better multiprocessor capabilities will help.

The problem is that MySQL is "dead" before Apache reach it's limits. 

> 3. Some say that the performance problem may be related to the
> built-in F_FULLFSYNC fcntl which will ask the drive to flush all of
> its buffered data to disk.
> This has been questioned and at the OS level it is always on. it is
> possible to turn this off if the app allows it. However this is a
> tradeoff between speed and data integrity.
> (http://lists.apple.com/archives/darwin-dev/2005/Feb/msg00072.html)
> Written by the dev who wrote BeOS and now works for Apple - excellent thread.

Interesting but the Anandtech SQL (and my own performance problem) 
appears while simply reading the database. So it is another source of 
slow performance on writen not an explanation for the Anandtech 
slow tests. 

> 4. On the Mac OS X Server list many people ran tests and got much
> better results than the anandtech report - some on G4s! Dr Rolf
> Jansen pointed out that setting:
> /usr/sbin/sysctl -w net.inet.tcp.delayed_ack=0
> overcomes problems with cheap NICs with too small transfer buffers.
> He suggests that Anandtech "simply had very low network performance
> from their PC-Box (possibly with a noname NIC with a small transfer
> buffer) to the Dual G5 Server."

I will try that. Can you give me the source please so I can learn a bit more. 

> 5. The inimitable Dan Shoop (also on OS X server list) acidly points
> out that MySql isn't enterprise level and that PostGreSQL is muuuuuch
> better. 

My objective is to use existing scripts specifically written for php/mysql. 
Switching to Postgress is not a solution in my case. 

> 6. The results - OS X may be a slower, but not 10x as reported by
> anandtech and your data integrity is better. Some test run a year ago
> by PC magazine give an entirely different picture:
> http://www.pcmag.com/article2/0,1759,1637655,00.asp

They test Apache, ou problem is MySQL. 

OS X slower than Linux is not my problem (I can wait a little bit more ;-)

The real problem is that after a certain amount of simultaneous
sessions on a php/mysql website (phpnuke, wordpress, wikipedia,
drupal, ...), the whole server completely "collapse" and start serving
verrrrry slowly (minutes per page) because of MySQL crawling to clear
it's queue of requests.

Until we get a different version of MySQL for MacOS X that use an
internal threading/forking code instead of relying on OS X (I
understood that was the approach of Oracle), a temporary fox could be
to cache phph/mysql request with eAccelerator 0.9.2a or Turck MMCache
2.4.6.

It would be great if Tenon would bundle one of those GPL php/mysql
caching solutions in iTools.

It must be possible as MAMP (Macintosh, Apache, Mysql and PHP) did it: 
http://www.mamp.de/en/home/

Hope it helps
-- 
Brice Le Blevennec, Digerati, ListDad, <http://www.brice.net> 
Emakina S.A. - The e-business Agency <http://www.emakina.com> 
Ex Machina Television S.P.R.L.- Television & Radio Production 
<http://www.cybercafe.tv> & <http://www.singlefm.com> 
Contact Office Group S.A. - The Professional Online Office 
<http://www.contactoffice.com>
---------
Tenon Intersystems' iTools Mailing List
To unsubscribe: send mail to
itools-request@xxxxxxxxxxxxxxx
with the body only containing:
unsubscribe
Find the searchable mailing list archives
at:
http://www.tenon.com/lists/html/iTools/