I started this post while Paul and I were in the TechEd Bloggers Lounge… which, from the number of folks “stopping by” must be more of an online thing :). We did have lots of folks visit with us in the DAT “Green” area but in the bloggers lounge, I thought I’d blog (maybe that’s the point?!). But, as luck would have it, I got side tracked at the end and ended up on the Women In Technology panel, then chatting with a customer, and then another… and, well, now it’s Friday and time to pack up and head home. This year’s TechEd was a combination of crazy busy highs with a very “where is everyone” lows. I asked about YOUR opinion of the split for TechEd this year and I think folks are waiting for the event to complete before they form opinions but for me, I felt like the event was smaller…without as much buzz as in TEs past. I’ve always really viewed TE as “the biggest US event of the year” where all of the buzz was created. It’s where I often learn about new (usually outside of SQL Server :) technologies and where I hear some of the latest buzz. This year, I just didn’t feel as though there was as much buzz (or as many people). And, I guess I’ve complained about how crowded it’s felt in years past so I guess both have their ups/downs – but, overall, TE just didn’t (in my opinion) have the same energy of TEs past… I’d definitely like to hear more opinions on this. Overall, it was nice to catch up with a few of the usual suspects.
As for a trip (and resources/tips/tricks) report… here I go :)
Monday was our “Overview of SQL Server 2008″ precon and that went really well… we were completely exhausted by the end of the day but we had a lot of great feedback. The general feedback was that it was the perfect way to start the week as it allowed people to get a really good more-than-just overview of the new features and help to determine which topics/sessions might be the most relevant for folks to attend. Also, I learned about a Performance Dashboard bug here that I thought I’d pass on as well - it’s not likely to happen to everyone and it’s related to a potential timeout problem. I didn’t find too many references to this problem though so if someone does have a more detailed link – let me know. The end result (a gentleman named Leif sent this script to me) is that you need to tweak the C:\Program Files\Microsoft SQL Server\90\Tools\PerformanceDashboard\setup.sql file so that the stored procedures deal with the timeout correctly. The change is to line 276:
sum(convert(bigint, datediff(ms, login_time, getdate()))) – sum(convert(bigint, s.total_elapsed_time)) as idle_connection_time,
sum(convert(bigint, CAST ( DATEDIFF ( minute, login_time, getdate()) AS BIGINT)*60000 + DATEDIFF ( millisecond, DATEADD ( minute, DATEDIFF ( minute, login_time, getdate() ), login_time ),getdate() ))) – sum(convert(bigint, s.total_elapsed_time)) as idle_connection_time,
Tuesday, I delivered my Indexing session “Are your indexing strategies working?” and the group really enjoyed some of the things I focused on – from DMVs to DTA to SSMS to Performance Data Collection in SQL Server 2008… there are many aspects to indexing and depending on your environment some are more relevant/important than others. There are definitely a lot of things to talk more about here… INCLUDE columns, statistics… In the interim, I’ve posted all of my demo scripts to our past conferences page here.
Wednesday, Paul delivered his Corruption session and it was excellent. He showed a variety of different situations that all have different workarounds – as well as discussed a few that don’t. He’s posted a few things about these sessions and he’s going to slowly detail each of his demos in blog posts.
Thursday, we delivered a Database Maintenance session which was a “Top 10″, per se. This was a great session for us but I wish we had done it as our first session instead of our last. I felt like this session was a wider session (in terms of topics) and our other two were deeper sessions (in terms of technical content)… But, it was still great fun to do. The thing that made it the most fun for me is that I decided at about 5pm the day prior that I wanted to tweak my demos a bit… and, 14 hours later, I was done. I decided that I would create two copies of a database – one for each of two laptops and then I would work really hard to create a “best practices database” and a “worst practices database”. What was most stunning to me was that the cumulative effect of all of the worst practices was actually a lot worse than I had thought it would be. In the first test run I did on stage my best practices database (before the index maintenance) ran at about 3500 measured statements per second. My worst practices ran at about 71. I absolutely did not expect that large of a difference… and, one of the things I’m truly looking forward to doing is breaking each of the worst practices down and comparing the breakdown of each to find what has the greatest impact.
The week was a big success overall and it was great to see a few folks that we seem to only see at these types of events……. see you next year!