MySQL Performance 6.0rc1
Marian Hettwer
MH at kernel32.de
Wed Oct 26 02:14:37 PDT 2005
Hi Folks,
I'm using RC1 on some of my test servers and workstations. So far, it
runs very good (as in stable and fast) :)
However, I thought of doing a small benchmark of MySQL in comparison to
Linux 2.6.8 (Debian Sarge) and the results are kinda... shocking.
On FreeBSD I took a pretty much generic kernel, I just removed debugging
options, added SMP and changed the scheduler from BSD to ULE.
I used super-smack to do a select-key test.
MySQL 4.1 was build from ports with BUILD_OPTIMIZED and BUILD_STATIC
super-smack gave me these numbers:
Query Barrel Report for client smacker1
connect: max=15ms min=1ms avg= 12ms from 10 clients
Query_type num_queries max_time min_time q_per_s
select_index 200000 2 0 3673.36
My Kernel Config:
http://unixoid.de/rabauke/freebsd/EDGECRUSHER
My dmesg:
http://unixoid.de/rabauke/freebsd/dmesg.freebsd6rc1
Since I have an identical second disk in my Dual Xeon BOX, I installed a
Debian Sarge with Linux SMP Kernel 2.6.8, installed mysql-server41
(which is 4.1.11) and did the same super-smack test. Results:
Query Barrel Report for client smacker1
connect: max=35ms min=0ms avg= 9ms from 10 clients
Query_type num_queries max_time min_time q_per_s
select_index 200000 0 0 14736.23
Whut? 14700 queries/second compared to only 3600 q/s on FreeBSD?
I do know, that rc1 is not about performance, this will be -RELEASE
then. I do know too, that it was just a quick shot and probably I should
compile my own version of the MySQL server from sources on FreeBSD and
Debian to have the very same mysql version.
However, I can't imagine that I would gain 4 to 5 times the performance
on FreeBSD just by compiling mysql myself...
Any Ideas where this huge difference comes from? Did I oversee some
debugging features still active in RC1 ?
Best regards and keep up the good work on 6.x!
- Marian
More information about the freebsd-current
mailing list