First two days of PASS Summit 2015 are now almost over. I just got back to the hotel from the welcoming party and decided to write a quick summary of these days. The pre-con sessions I attended to were “The Complete Primer to SQL Server Virtualization” by David Klee, Argenis Fernandez and Jimmy May on day one and “The Enterprise Scripting Workshop” by Sean and Jennifer McCown on day two.
When it comes to virtualization there’s been so much improvement in the hypervisors and technology in recent years that have minimized the virtualization overhead on performance, that I’ve come to view it as a very serious option for running production SQL Servers. And listening to the real life examples of huge environments running virtualized Tier-1 servers and the benefits in terms of cost optimization (both licensing and hardware) as well as in building resilient infrastructures, I find it difficult to come up with sensible counter-arguments to virtualization.
There are of course some challenges there, but that is always the case when building and designing complex server infrastructures. The good news was that we’re already doing many of the things in our organization that are required for succesfully deploying virtualized server environments, such as benchmarking and testing and having an understanding of the workloads involved. And while we already have a solid foundation to which to build on there were some new ideas and concepts, especially in area of high availability and disaster recovery that we need to look more into.
The scripting workshop touched a lot of topics I have worked with recently, like standardized SQL Server deployments as well as maintenance and administration routines. However one of the more eye-opening topics, at least for my, had to do with either being busy or being productive at work. There were so many familiar examples of trivial tasks that could and definitely should be automated to free up time for productive work.
There were also lot of good guidelines on how to build a scripting solutions for administration that included adding things like error handling and logging into them. There were also lot of scripting examples that helped to drive the point home. And just like with the virtualization pre-con, I was very satisfied to see that we are on correct path again and I now have much more knowledge on how to improve the work we’ve already done.
We went to the welcoming party but didn’t really stay too long, just enough to get some food and couple drinks before returning to hotel. We did meet a person from Texas who had a Finnish friend who now worked in Texas also. Coincidentally him and the Finnish friend both worked in oil industry, which I suppose is very stereotypical Texan thing 🙂
We’ll be heading back to Conference Center at 7AM tomorrow for some breakfast and then to Day 1 Keynote. My last session ends at 6PM, so it’ll be a long day full of learning.