April 18, 2014

HighLoad 2007 Review

Almost two weeks have passed since HighLoad 2007 conference in Moscow, Russia so I’m pretty late with feedback and my only excuse I was pretty busy to spare some time for it.

Lets start with bright side.
The conference had a lot of great sessions by experts in their field. A lot of top Russian companies such as Yandex Rambler, Mail.Ru had sessions covering their internal architecture. Speakers were very open in many cases sharing data about their internal implementation, for example Yandex would tell you a lot more details about their search process than Google would be willing to share. I specially mention Yandex here as they had very well presented team on the conference and had largest number of sessions presented by the same company.

The other very impressive thing was publishing of materials for previous (Russian Internet Technologies) conference in the book and distributing them among this conference attendees. And it was not just slides as I’ve seen on other conferences – the talks from previous conference were recoded and them converted to the text and put together with slides. I can only imagine how enormous effort it required but it is very handy to get.

Also the conference had all sessions recorded and broadcasted on RuTube – this however did not always work well. As my wife tells me my sessions were not broadcasted for some reason and also I did not find them on the Session Videos page.

The presentation stand was prepared much better this time. You still was requested to use conference laptop rather than your own but at least you had remote control allowing you to switch slides manually rather than ask for slide to be changed as on previous conference.

Generally I had very enjoyable time talking to speakers as well as a lot of conference attendees. We had a good talk with Bruce Momjian and I learned a lot of interesting low level details about PostgreSQL transactional engine internals.

Now what I did not like.
The previous conference was rather crowded and this one looked like larger one but taking place in much smaller space. I would say it was about half the size of MySQL Users Conference 2007 but probably had 1/8 of the space.
This means corridors were very crowded (forget about place to sit outside of speaker rooms) and good sessions were so crowded you could not get even standing place.

I also did not like the lunch idea – to make it less crowded break was split into 2 parts – the break in one room was when other had session going which meant you have to choose between food for your stomach or for your brain. Plus as I mentioned lunch was very crowded with no place to sit.

Registration was the other problem. Registration opened at 9 and sessions were supposed to start at 10. This produced huge crowd fighting to get to registration desk. A lot of people (including myself) also could not be found in the list and had to go to special desk to resolve the issue which had a long line as well. Were were no special speaker registration as well.

Another problem worth to mention was schedule. Because registration problem sessions went late and other issues agenda changed from printed one, while were were no board or anything which would show current schedule. Were were also time swaps which were not announced widely enough so few times I came to the session just to find the session was changed or actually moved earlier so I already missed it. In my case of spending a lot of time talking to the people rather than just visiting sessions up to date schedule information is important.

Different
Few things for this conference were just different from most of International conferences I’ve been to.

First – the amount of interest you raise is enormous – I spent 2+ hours after each of my sessions answering questions, not to mention questions/talks you get at other time. And this was same for many other speakers.

Second – it was very interesting to see probably every 3rd person on the conference out where hunting for people – good engineers are extremely hard to find in Russia now as the market is pretty hot. A lot of people I spoke to were approached about job opportunities so it was great place for job hunting as well.

Third – It was truly Russian conference. Were were two sessions in English – once from Microsoft, another by Bruce Momjian from PostgreSQL which were translated to Russian live (I did not listen Russian so have no clue about translation quality). There were also people from Oracle speaking on the conference but this is about it. It was very surprising to see no Google people attending, even though they now have office in Russia.

In generally it surely was worth the visit for me and I think this conference has potential of tuning to very exciting yearly event.

P.S My sessions from the conference uploaded to MySQL Presentations section as usually. Note – they are in Russian.

About Peter Zaitsev

Peter managed the High Performance Group within MySQL until 2006, when he founded Percona. Peter has a Master's Degree in Computer Science and is an expert in database kernels, computer hardware, and application scaling.

Speak Your Mind

*