Two weeks ago was a busy one. I helped coordinate and spoke at our second SQL Saturday Omaha on April 6, flew to Las Vegas for the new SQLintersection conference at the MGM Grand, and then flew to Chicago to speak at SQL Saturday Chicago.
SQLintersection is part of a group event trifecta along with DevIntersection and AngleBrackets. Take a look at the following speaker lineup. How could you NOT want to attend these conferences! MVPs, industry founders and executives, and legendary experts abound!
(This list is not all inclusive, and for those speakers that I omit, it’s not because I don’t like you! It’s because I’m mainly putting the speakers whose sessions I attended)
- Aaron Bertrand (b | t)
- Connor Cunningham (b)
- Kevin Farlee (b)
- Grant Fritchey (b | t)
- Steve Jones (b | t)
- Andrew Kelly (b | t)
- Kevin Kline (b | t)
- Kendra Little (b | t)
- Brent Ozar (b | t)
- Jeremiah Peschka (b | t)
- Paul Randal (b | t)
- Joe Sack (b | t)
- Paul Sheriff (w)
- Kimberly Tripp (b | t)
- Bob Ward (b | t)
- Paul White (b | t)
First, on Monday I decided to attend a preconference session. I chose Paul Sheriff‘s pre-session called ‘Application Architecture for Multiple UIs’. Are you thinking that this session is a strange choice for me? I decided to branch out a bit and see how modern (read: mobile, cloud, etc.) applications are handling data access methodologies and see how I could grow my application performance triage methodologies. He did a fantastic job of talking through the various tiers of modern apps and demonstrating how each interact and relate.
Monday evening was the opening keynote presented by Tim Huckaby on ‘The Engaging User Experience and the Natural User Experience.’ Wow. Use a Kinect motion sensor to allow physical therapy exercises to be performed at home and effectively graded to ensure it is being done each day. Use a Surface table to plan a wedding. I love solid applications of emerging technology. He showed fun and practical uses for the gigantic Surface table, Kinect sensors, and demoed emerging UIs.
OK – On to the technical content!
Tuesday Morning Joe Sack kicked off the day with an amazing session on CPU Performance Issues. Joe, you are amazing. He demonstrated remediation techniques for common CPU-based performance problems. Every scenario he demonstrated are situations that I encounter in the field when doing performance troubleshooting, and his triage process is something that every DBA should learn from.
Next, Paul Randal presented two sessions on ‘Making SQL Server Faster’. I have heard portions of these two sessions before during the SQLskills Immersion Events, and these items are part of my checklists when I do my performance work for customers. Every tip and trick in these two sessions are invaluable and should be considered part of your processes.
Next, Kevin Farlee gave me a crash course in Project Hekaton, a new in-memory component of the next version of SQL Server. I cannot speak highly enough of Kevin and Hekaton. Microsoft is building Hekaton to address some of the internal bottlenecks within the core SQL Server engine, and I cannot wait to get my hands on the product (whenever that may be) so that I can start learning. The applications of this technology are wide and the implications are many.
Next, Kevin Kline and Sumeet Bansal discussed how SQL Server AlwaysOn and Fusion-io helped a customer to achieve extremely high transactional throughput on SQL Server 2012 while adding rock-solid HA and DR at the same time. Their findings were interesting. At large scale, they recommended isolating one CPU core per 16 cores for a dedicated processor for parsing network traffic and I/O processing. They were able to sustain hundreds of thousands of IOps underneath AlwaysOn workloads running on Fusio-io PCIe SSD cards, and the SQL Sentry tools scaled with the environment with ease.
How’s that for an official first day of the conference!
Wednesday started off with Brent Ozar focusing on one of my favorite topics – ‘Oh No! They’re making me virtualize!’ By now you all know that I am an advocate for 100% virtualization of the datacenter, and speak and write about it frequently. While the presentation took a different twist than how I normally approach it, the approach was fresh and very solid. The end result is the same as mine – know your workload, clean it up before you virtualize it, and you will probably have a good experience. Great job!
Next, Joe Sack was up again to change gears and focus on the identification and remediation of common performance scenarios. The situations covered were so wonderfully real-world that I started remembering different projects that fell under each of his presented situations. Again, learn his approach. It’s wonderful!
Next, Connor Cunningham talked about ‘Architecting Large-Scale Services on Azure’. I have not had too much experience with Microsoft Azure SQL Databases to date, but after seeing his session, I am very eager to get started and ramp up. I learned quite a bit of tips and tricks about the Azure architecture, items that will need to be adjusted within existing apps to provide an optimal user experience, and solid techniques for architecting new applications for scalability. This service looks very useful, and I love watching this platform evolve.
At this point the brain is going into overdrive on how I want to start rethinking some of the methods that I use to incorporate everything that I have learned… who needs sleep, eh?
Thursday was also great. Bob Ward discussed solid methods for ‘Windows Azure SQL Database Troubleshooting’, which continued my deep dive on Azure for the conference. Bob, known for his >500 level sessions at other conferences that I have attended, toned it down so us mere mortals with little Azure experience could digest it on the first gulp. I learned even more about Azure and how to better manage it. Again, I cannot wait to dive in headfirst!
Next, Bob launched straight into a second session on ‘SQL Server 2012 Database Engine Tools and Tips’. Again, wow. I am taking this presentation (OK – actually I’m taking ALL of the presentations) back to my crew at House of Brick so we can absorb as much as we can.
Next, Andrew Kelly gave me a session on ‘SQL Server 2012 Query and Procedure Call Statistics’. Anyone dealing with performance tuning (and that should be ALL DBAs, right?) should consider every shred of knowledge from Andrew (and all of the other speakers at this conference for that matter) like gold and follow every word they publish.
OK, I know you’re thinking that I’m just gushing at this point, but I am really excited about the potential of this conference as time goes on. The speakers and content in this conference were absolutely top notch, and I am thankful to have been able to attend. I learned a great deal on some bleeding edged topics, and I cannot wait until the next conference!