I had the opportunity and privilege to present Introduction to SQL Server Profiler to the Tampa SSUG last night. I was impressed by the turnout as it was a full house with about 40 in attendance. I arrived early, and had a chance to meet a few people, and attempt to work on my networking skills. I learned I need more practice as I’m still having trouble remembering names of people that I met for the first time. Some of the people I met were Jonathan Kehayias, @jmkehayias on twitter; Carlos; Andres (I hope I spelled it right); @SQLChicken, Jorge Segarra; Ron Dameron and Nathan (Fraggle on SSC), who I had met at SQLSaturday 8 – Orlando last fall; and, of course, Pam Shaw (@pamshaw) who does a great job leading the group.
After Pam went through the announcements Jorge led a discussion on the pros and cons of social networking (Twitter, LinkedIn, Facebook, etc…). Some good stories were told about how problems had been solved through the active SQL Server community on Twitter, and a couple of attendees shared how that they had used LinkedIn to either find a job or a candidate for a job. I think some attendees may have been convinced that there are business/professional reasons to use social networking tools.
After that discussion I was up with my presentation on Profiler. This was the first time I’ve done the presentation on my new laptop with SQL Server 2008 installed so I had a couple of minor technical glitches like not having reporting services running as I expected and adjusting to how Vista handles presentation mode. I still need some work on that. I always start by asking who has used or uses Profiler and I am always surprised that every hand doesn’t go up. The other question I ask is how many people know that SQL Server 2005/2008 has a server-side trace running (the Default Trace) on install. Again, the majority of attendees do not know this. Overall I covered everything I planned in time and had some good questions around the difference between SP:Completed/RPC:Completed and SP:StmtCompleted/RPC:StmtCompleted and good discussion about why you even have the option of having Profiler write to a file or table while running the GUI, since it is recommended against and you can save to a file or table when the trace is done. Jonathan Kehayias gave me a couple of tips on things I missed or should have said, like when tracing lock events DON’T trace Lock:Acquired as that will fill up a trace fast. Remember locks are good, deadlocks are bad. It felt like everyone was pretty happy with the information presented and several people let me know they enjoyed the session, so I guess it well.
Even though it went well, I know I can do better, but the more I present the better I’ll get. I’ve got SQLSaturday 14 – Pensacola coming up on June 6th where I’ll be doing my Profiler presentation and Space Coast User Group in September where I’ll be doing a presentation on mining the Default Trace. I hope to see you at one of the upcoming events!