Tim is now vocal

Tim at the datacenter
Tim is one of most humble and intelligent developers I’ve ever met – and we’re extremely happy having him at Wikimedia. Now he has a blog, where the first entry is already epic by any standards. I mentioned the IE bug, and Tim has done thorough analysis on this one, and similar problems.

I hope he continues to disclose the complexity of real web applications – and that will always be a worthy read.

Knol

There isn’t much to talk about Knol technology – it is either nicely engineered or missing (they probably thought that search is main tool for collaboration). Of course, many issues are already covered by others, but…

My first look was at the featured articles. What was wrong?

  • It features ‘closed collaboration’. Actually, thats no different from a blog, then…
  • It doesn’t care much about the licensing – featured articles had images with “all rights reserved”, or images taken from Wikipedia, with attribution but without share-alike clause. Also, no share-alike license forbids importing of content from many other places, but as we see it – nobody cares. ;-)
  • It doesn’t care about linking. Google search was based on the web links. Wikipedia was built on top of lots of broken links (oh, and working ones too). And nobody is going to type a Knol URL.
  • It doesn’t seem to have community tools. It just doesn’t.
  • WYSIWYG editing leads to articles without structure, just some text parts bolder than the other.

So for now, it seems to be pure-engineering approach at the problem, without looking at actual work done, social implications or properly respecting copyrights.

One needs community for that. Community helps not only with content, but with style, metadata, organizing, and most of all – ensures that project maintains values and spirit.

Wikipedia at Velocity conference

Next Monday I’ll be presenting (if jetlag doesn’t kill me) at Velocity 2008 – webops and performance conference. It won’t be my first time talking about Wikipedia infrastructure, but this time people will know the technology and scaling methods anyway.

As I see it, in such context Wikipedia is more interesting as a case of operations underdog – non-profit lean budgets, brave approaches in infrastructure, conservative feature development, and lots of cheating and cheap tricks (caching! caching! caching!).

Also, I’ll be able to share (making audience jealous) how it is great to be on non-profit ops team (and one of example perks – we can be cheap about getting conference passes too ;-)

The best part (for audience, not for me) – I will be forced to be honest. Nearly whole tech team will be at the event, and if I fail to attribute any developments, or start talking crap – not only they can throw rotten tomatoes, but also disable my login access and claim they never knew me, without me being able to fight back :) I didn’t publicly present in front of these guys since 2005 – will be tough.

5.0 journal: various issues, replication prefetching, our branch

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. :)

Google does encyclopedia: Knol

It is all closed yet, announcement by VP Engineering tells us Google is launching their idea of encyclopedia – looking for people who can write authoritative articles. No words on licensing apart from “we want to disseminate it as widely as possible”, though author-centric view is more what Citizendium, than Wikipedia wants to do.

Ad revenue sharing poses many interesting questions, especially in collaborative effort. As Wikipedia now provides page view statistics, Google (or knollers) may just work on top of the cream pages (by knowing search trends), and have very distorted overall content. Now it is closed, invite only, so we can’t tell anything more. Time will show. Good to know more organizations are believing about aggregating and disseminating knowledge – it is Wikipedia’s mission, and it is nice to have partners. :-) Though of course, there might be some tensions with Search Quality team…