First of all, I have to apologize about some of my previous remark on 5.0 performance. I passed ‘-g’ CFLAGS to my build, and that replaced default ‘-O2’. Compiling MySQL without -O2 or -O3 makes it slower. Apparently, much slower.
Few migration notes – once I loaded the schema with character set set to binary (because we treat it as such), all VARCHAR fields were converted to VARBINARY, what I expected, but more annoying was CHAR converted to BINARY – which pads data with bytes. Solution was converting everything into VARBINARY – as actually it doesn’t have much overhead.
TRIM('' FROM field) eventually helped too.
The other problem I hit was paramy operation issue. One table definition failed, so paramy exited immediately – though it had few more queries remaining in the queue – so most recent data from some table was not inserted. The cheap workaround was adding -f option, which just ignores errors. Had to reload all data though…
I had real fun experimenting with auto-inc locking. As it was major problem for initial paramy tests, I hacked InnoDB not to acquire auto-inc table-level lock (that was just commenting out few lines in ha_innodb.cc). After that change CPU use went to >300% instead of ~100% – so I felt nearly like I’ve done the good thing. Interesting though – profile showed that quite a lot of CPU time was spent in synchronization – mutexes and such – so I hit SMP contention at just 4 cores. Still, the import was faster (or at least the perception), and I already have in mind few cheap tricks to make it faster (like disabling mempool). The easiest way to make it manageable is simply provide a global variable for auto-inc behavior, though elegant solutions would attach to ‘ALTER TABLE … ENABLE KEYS’ or something similar.
Once loaded, catching up on replication was another task worth few experiments. As the data image was already quite a few days old, I had at least few hours to try to speed up replication. Apparently, Jay Janssen’s prefetcher has disappeared from the internets, so the only one left was maatkit’s mk-slave-prefetch. It rewrites UPDATEs into simple SELECTs, but executes them just on single thread, so the prefetcher was just few seconds ahead of SQL thread – and speedup was less than 50%. I made a quick hack that parallelized the task, and it managed to double replication speed.
Still, there’re few problems with the concept – it preheats just one index, used for lookup, and doesn’t work on secondary indexes. Actually analyzing the query, identifying what and where changes, and sending a select with UNIONs, preheating every index affected by write query could be more efficient. Additionally it would make adaptive hash or insert buffers useless – as all buffer pool pages required would be already in memory – thus leading to less spots of mutex contention.
We also managed to hit few optimizer bugs too, related to casting changes in 5.0. Back in 4.0 it was safe to pass all constants as strings, but 5.0 started making poor solutions then (like filesorting, instead of using existing ref lookup index, etc). I will have to review why this happens, does it make sense, and if not – file a bug. For now, we have some workarounds, and don’t seem to be bitten too much by the behavior.
Anyway, in the end I directed half of this site’s core database off-peak load to this machine, and it was still keeping up with replication at ~8000 queries per second. The odd thing yet is that though 5.0 eats ~30% more CPU, it shows up on profiling as faster-responding box. I guess we’re just doing something wrong.
I’ve published our MySQL branch at launchpad. Do note, release process is somewhat ad-hoc (or non-existing), and engineer doing it is clueless newbie. :)
I had plans to do some more scalability tests today, but apparently the server available is just two-core machine, so there’s nothing much I can do on it. I guess another option is grabbing some 8-core application server and play with it. :)
8 thoughts on “5.0 journal: various issues, replication prefetching, our branch”
Great, i learn more with your blog than reading 10^2 fucked manuals of mysql.
And the most important, i can see all data benchmark to get a real aproximation to know how mysql runs and his limits.
Thanks a lot for this post and for your blog ;-D
There is mutex contention on the InnoDB tx log mutex (log_sys->mutex) that limits SMP speedup for tests such as your (reloading a database). We don’t have a patch for it yet.
Yeah, log_sys->mutex shows up a lot, though now I seem still to hit bufpool giant mutex.
Please, share the hack to parallelize mk-slave-prefetch. Attached as a patch to a bugreport would be great.
I agree with you on the problems with only one index, etc etc — in specific cases, this may be possible to solve, but in general is impossible to solve outside the server IMO.
Baron, patch was linked by the post, here it is as a plain url: http://dammit.lt/snippets/quick-mkprefetch-hack.txt ;-)
Regarding the rewrite of UPDATE to SELECT (a.k.a., the oracle algorithm presented by Paul Tuckfield). I actually have a feature clone involving adding scripting support to the replication threads, so that the oracle algorithm can be written in something like 20 lines of Lua.
We are in the middle of setting replication interfaces for Marc’s and Wei’s semi-sync patch, so once that is done, I can adapt the feature clone to use that interface instead.
On rereading this again I noted that you’re using -O2. Please consider using -O3, since function inlining (-finline-functions) is not turned on by default for optimization levels below 3, and that flag is a big win most of the time. Of course, you can also add that flag explicitly.
Mats, our in-house builds force -O3 nowadays :)
Comments are closed.