Hotsos Day 3.07.0

Up again at 4am, then, and a frantic dash to finish the presentation. Most of it was just pacing, transitions, jokes etc. but it still takes a *lot* of time. I thought I might attend a presentation or two before my own and really aimed for that but in the end it was the usual no breakfast/no presentations lead up to my own. Just time for a couple of smokes outside and then it was on with the presentation.

I think it went pretty well and judging by the number of follow-up questions and discussions, people are certainly very interested in Parallel Execution and hit their own unique problems as well as a couple of shared ones.

I kicked off with the usual survey of how many people read this blog and there were at least two or three times as many as last year (in a much smaller audience). They weren't so keen to stick their hands in the air for the obligatory photo, though, but the Oracle Sponge had enough warning to put his shades on for his picture. (He had no advanced warning, I assure you, and I didn't even notice it until he pointed it out later). He's looking very suave and sophisticated these days - must be that move to Virginia! (And, yes, there are quite a few mates in the front row (L-R) - Mark Brady, Alex G, David Aldridge and Shrek. I've spent a lot of time with those guys this week. You might just be able to recognise them in the fuzzy photo. Patty is flashing her winning smile from the second tow.)

I was a little concerned about how the content would go down because some of the monitoring approaches I was suggesting were decidedly old-school and far from YAPP or Method R, but if time spent in the o/s run queue is unaccounted-for in those approaches, then needs must. As the presentation progressed I sensed a good reaction from the room, though.

I was particularly happy with how the presentation aspects went. Some days you just feel on form and this was one of those occasions. If I'm not mistaken you might see a photo or two from Alex later that will take some explaining! (Oh, and thanks to Mark and Alex for holding up the Ice Skating judge '5' signs at the end )

A big shout out to Myron Wintonyk who, during the presentation, suggested the idea of having different applications connect to the database through listeners started up under different operating system accounts. The idea being that you can then use o/s tools that can investigate activity at the o/s account level. We were talking about DTrace at the time, but I think the same principle would hold true for several uses as we were discussing earlier today.

After the presentation, I was answering follow-up questions for 30 minutes. Alex G was too polite to desert me, so he was late for his lunch too. Cheers, mate (Particularly as it was fajitas - a firm Gorbachev favourite - and with guacamole this time!)

Definitely a good morning and the worrying is over for another conference

Now I'm in the final event of the week, the afternoon training session - "Diagnosing and Avoiding RAC Performance Problems" - which I definitely don't want to miss, but I'll catch up on more later.