mysql performance on 4 * dualcore opteron

Steven Hartland killing at multiplay.co.uk
Wed Apr 5 17:04:16 UTC 2006 

Looking at this on a dual box here ( waiting for the new MB for dual dual core )
All the time is spent processing super-smack and only 25% on mysqld.
Even dropping to 10 clients a large portion is take by the clients.
That said there is a lot that can be gained by using the tweaks out there
i.e. ULE + libthr + TSC + context_time.patch + cpu_acct_1.patch + cpu_acct_2.patch 

Adding these jumps from a baseline:
select_index    2000000 8       0       18624.60
to:
select_index    2000000 5       0       29942.10

The biggest increases coming from libthr ( thanks DavidXu ) and the ULE
scheduler.

[log]
== 4BSD + libpthread + ACPI-Fast ==
super-smack -d mysql select-key.smack 100 10000
Query Barrel Report for client smacker1
connect: max=46ms  min=6ms avg= 25ms from 100 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 8       0       18624.60

super-smack -d mysql select-key.smack 10 100000
Query Barrel Report for client smacker1
connect: max=5ms  min=0ms avg= 1ms from 10 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 0       0       23983.87

== 4BSD + libthr + ACPI-Fast  ==
super-smack -d mysql select-key.smack 100 10000
Query Barrel Report for client smacker1
connect: max=107ms  min=2ms avg= 45ms from 100 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 13      0       22413.39

super-smack -d mysql select-key.smack 10 100000
Query Barrel Report for client smacker1
connect: max=2ms  min=1ms avg= 1ms from 10 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 0       0       26841.07

== 4BSD + libthr + TSC ==
super-smack -d mysql select-key.smack 100 10000
Query Barrel Report for client smacker1
connect: max=46ms  min=1ms avg= 21ms from 100 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 11      0       23428.03

super-smack -d mysql select-key.smack 10 100000
Query Barrel Report for client smacker1
connect: max=2ms  min=0ms avg= 1ms from 10 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 0       0       26403.95

== ULE + libthr + TSC ==
super-smack -d mysql select-key.smack 100 10000
Query Barrel Report for client smacker1
connect: max=41ms  min=0ms avg= 23ms from 100 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 5       0       28581.18

super-smack -d mysql select-key.smack 10 100000
Query Barrel Report for client smacker1
connect: max=4ms  min=0ms avg= 1ms from 10 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 0       0       30128.44

== ULE + libthr + TSC + context_time.patch + cpu_acct_1.patch + cpu_acct_2.patch ==
super-smack -d mysql select-key.smack 100 10000
Query Barrel Report for client smacker1
connect: max=27ms  min=0ms avg= 14ms from 100 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 5       0       29942.10

super-smack -d mysql select-key.smack 10 100000
Query Barrel Report for client smacker1
connect: max=12ms  min=0ms avg= 4ms from 10 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 0       0       31057.52

== 4BSD + libthr + TSC + context_time.patch + cpu_acct_1.patch + cpu_acct_2.patch ==
super-smack -d mysql select-key.smack 100 10000
Query Barrel Report for client smacker1
connect: max=54ms  min=20ms avg= 38ms from 100 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 9       0       24144.22

super-smack -d mysql select-key.smack 10 100000
Query Barrel Report for client smacker1
connect: max=2ms  min=0ms avg= 1ms from 10 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    2000000 0       0       27073.46

** update test **
super-smack -d mysql update-select.smack 10 100000
Query Barrel Report for client smacker
connect: max=3ms  min=0ms avg= 0ms from 10 clients 
Query_type      num_queries     max_time        min_time        q_per_s
select_index    1000000 1       0       6468.70
update_index    1000000 0       0       6468.70
[/log]

Machine:
Dual 244, 2Gb running FreeBSD 6.1-PRERELEASE (i386)
Package install of mysql 4.0
Port install of super-smack

Notes:
No detectable disk activity thoughout the tests
ULE scheduler breaks the output from top with everything showing as
WCPU 0% in the 100 concurrency test and the numbers not adding up
at all in 10 concurrency test or showing 0%.
To get context_time.patch to work I needed the attached patch which
is basically two failed chunks of: kern/kern_exit.c moved to kern/kern_thread.c

    Steve
----- Original Message ----- 
From: "Sven Petai" <hadara at bsd.ee>
To: <freebsd-performance at freebsd.org>
Sent: Tuesday, April 04, 2006 5:42 PM
Subject: mysql performance on 4 * dualcore opteron


> hi
> 
> Before I begin, let me just say that I'm probably aware most of the threads 
> about mysql performance in various fbsd lists over last couple of years, so 
> please let's not consentrate on the usual points made over and over again 
> like how filesystems are mounted under linux, how fast time() is or how 
> various combinations of scheduler/threding library/compiler flags give you 
> ~5-10% better performance. It's very unlikely that any of these reasons, or 
> even all of them together can explain performance differences of 2-3 * 
> 
> so now a little bit of the backround...
> I usually use MySQL benchmark called super-smack as one of the benchmarks on 
> all the new machines to get a general feeling of the servers performance.
> I certainly agree that the default smack workloads are far too simple to say 
> much about actual production performance, but still... better than nothing...
> 
> In general 2.4Ghz amd64 UP box (6.1 betaX) can do about
> 17400 q/s with select-smack+4bsd+thr combination and
> 4300 q/s with update-smack+4bsd+thr
> 
> on dualcore 2Ghz opteron (6.1 prerelease) the results are:
> 20000 q/s with select-smack+4bsd+thr and
> 4500 q/s  with update-smack+4bsd+thr
> 
> performance for update-smack seems to be always 4XXX q/s, no matter how many 
> CPUs the box has or what kind or raid controller/disks are used (i have 
> tested on about 8 rather different machines).  I have no idea if IO on all 
> the servers I have tried really maxes out at this point or is there some 
> bottleneck in UFS.
> select-smack performance gains on dualcore are not quite as good as one might 
> expect, but then again that dualcore box uses ECC memory which is probably 
> somewhat slower because of the checksum calculations, and synchronisation has 
> some overhead too... 
> Anyway all in all I'm more or less happy with these results, even though linux 
> will do about twise as much selects on the same hardware.
> 
> Today I had a chance to test 4 * 2Ghz dualcore opteron machine,  so this 
> machine has 8 cores in total and 8G of RAM.
> 
> Now, on that server I get:
> 11000 q/s for select-smack+4bsd+thr combination (with KSE it's around 6000 
> q/s, ule+thr gives somewhere around 12000 q/s)
> 4100 q/s for update-smack+4bsd+thr
> 
> So the 8 core machine got almost 2* worse result for select than UP server.
> 
> After some tinkering I found out that renicing mysqld to -5 will make it push 
> out 21000 q/s (4bsd, thr), so I suspect part of the problem is in the 
> scheduling - probably super-smack with it's 100 processes gets just a lot 
> more CPU time otherwise than mysql with it's 100 threads servicing them. 
> But anyway even this result is still only about equal in performance to what I 
> get from dualcore machine.
> 
> As I ran out of good (macro)tuning ideas at this point, and wanted to make 
> sure higher scores are indeed achievable, I tried Linux on the same hardware.
> Here are the results for same tests on Suse enterprise linux 9 
> (2.6.5-7.97-smp):
> 76857 q/s for select-smack
> 10050 q/s for update-smack
> 
> the mysql configuration was identical to the one I used under freebsd 
> (my-huge). 
> This Suse uses ReiserFS, but I have no idea about what kind of FS guarantees 
> it provides, didn't see any sync/async stuff in the mount output.
> I also repeated the tests on identical box that had Fedora installed 
> (2.6.9-22-ELsmp) and used ext3'fs.
> select-smack results were obviously almost the same as it doesn't touch the 
> FS, update was about 8000 q/s.
> 
> I'm relativelly sure that this kind of huge performance differences can't be 
> explained by mere speed difference of time(), I haven't yet tested phk'd and 
> roberts timer hacks, but at some point in time I rewrote mysql's timing code 
> to completelly avoid any calls to time() by keeping internal timestamp that 
> was updated from TSC reg. value. It was certainly very ugly and imprecise, 
> but worked well enough since mysql uses these code paths mainly for 
> statistics and for setting various safeguard timeouts. 
> Even with ~90% time() calls removed the performance still didn't get 
> measurably better.
> Of course it's possible that I fucked up somehow, so if someone has tested 
> roberts and phk's changes then it would be certainly nice to hear about your 
> results.
> 
> To make the long story short - does anyone have any good ideas about where 
> might the bottleneck and how to debug it ?
> 
> PS
> Here's some system/test information:
> super-smack was used with concurrency of 100 and reqs. set to 10000
> it was running on the same machine as the mysqld and connections were done 
> over local socket.
> 
> timer: acpi-fast in all the cases
> mysql: 4.1.18_2 from ports, table type is myisam
> mysql configuration file:
> http://bsd.ee/~hadara/debug/mysql3/2way/my.cnf
> in general it's just my-huge.cnf from mysql distribution, with increased 
> max_connections
> 
> kernel config is GENERIC-SMP (no it doesn't have WITNESS enabled)
> == 4 * dualcore opteron ==:
> vmstat 1, during select-smack test:
> http://bsd.ee/~hadara/debug/mysql3/8way/vmstat.txt
> dmesg:
> http://bsd.ee/~hadara/debug/mysql3/8way/dmesg.boot
> sysctl -a:
> http://bsd.ee/~hadara/debug/mysql3/8way/sysctl.txt
> 
> == 1 * dualcore opteron ==:
> vmstat 1, during select-smack test:
> http://bsd.ee/~hadara/debug/mysql3/2way/vmstat.txt
> dmesg:
> http://bsd.ee/~hadara/debug/mysql3/2way/dmesg.boot
> sysctl -a:
> http://bsd.ee/~hadara/debug/mysql3/2way/sysctl.txt
> _______________________________________________
> freebsd-performance at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-performance
> To unsubscribe, send any mail to "freebsd-performance-unsubscribe at freebsd.org"
> 
>

================================================
This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. 

In the event of misdirection, illegible or incomplete transmission please telephone (023) 8024 3137
or return the E.mail to postmaster at multiplay.co.uk.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: kern_thread.c.patch
Type: application/octet-stream
Size: 649 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-performance/attachments/20060405/b7e4b8e0/kern_thread.c.obj

More information about the freebsd-performance mailing list