So I’ve really slacked on blogging since starting my own website. My thought was with my own domain that I would do this more. I have a lot of partial drafts right now that are eager for publication.Well, hopefully I have motivating myself merely by typing that.
This past Saturday, I took part in a SQL Server 2016 Discovery Day in Raleigh-Durham, to learn about the new features in this year’s edition in a hands-on environment with some of the PASS data available to the public. It was a pretty solid way to spend a really hot weekend afternoon, geeking out over bells and whistles and data we don’t always get to access. First, our chapter leader (or as Kevin refers to himself, Grand Poobah) gave a presentation concerning all of the new features that made the 2016 edition the best one yet for all of us who spend a ton of time working within the Microsoft stack. I particularly looked forward to putting with the Transaction Performance Analysis Overview into action, along with checking out R Services. The second presentation was one on columnstore indexes delivered by MVP Rick Heiges (fun fact…he’s related to a former president of my alma mater, which is how we first made acquaintance some time ago). Indexes aren’t my strong suit, so I can only say that he showed us how this index focus will make it easier for folks like me to improve in that facet. Then it was time to form teams and start hacking it.
We were given some really dirty data sets (IMO, at least) and would use these to create analyses on geographical locations on SQL Saturday and Summit attendees, and correlations between the topics presented and virtual chapter membership. Our group (Team Cheetah) had a diverse background from database engineers to report writers. We really focused strongly on the PowerBI aspect, and finding ways to clean up some of the data related to tracks and zip codes/addresses. After the fact, I realized on my part that I could have used more columnstore index feature, but I’m guessing that the leaders liked my decision to use lookup tables to make the above graphic work. I think I got a bit excited explaining how we found that topics presented within an 18-24 month span were the most frequent of Summit presentations in the next year, going on…and on…and on about the dashboards and the graphic types. After the fact, our team found more we could have done with the hours we had, but considering the rookie status we felt alright with what we would present and be judged on (use of new features, dashboards, analysis, etc). So maybe we would get points for utilizing the presentation and talking about the muddied data we had. Plus, the engineers were good at the geospatial aspect and cleansing.
Well, our team won the contest, with gift cards as prizes, so obviously that was a cool feather in the cap.
— Kevin Feasel (@feaselkl) June 11, 2016
I will say that an exercise like this was rather interesting and cool, but I’m wondering what we could do for the PASS Community if we were to take this on for longer than an afternoon. If we had a team with BI Developers (like myself), ETL and data quality experts, analysts, and data scientists, then we could have some very useful information for trending. That may be the ultimate goal of this exercise, but it can become a stepping stone for a true hacking session.
Plus, this is an opportunity for a lot of the members of our local chapter to try their hand at civic hacking…myself included. One of my recent goals/fascinations has been civic hacking. The opportunity to take public data and improve a community using this information. In the Triangle alone we have many open data events, including CityCamp and Datapalooza. I’ve had a chance to sit in on some of the ideas and even add an idea in there, but a hackathon is something I’d like to try in full. Today gave me an idea on how it may work in a controlled environment, and what I could improve on when presenting findings. I would love to see more of this happen, and knowing that the community can have its own members help out with the solutions to make things better. Then, in turn, we can go out and use open data to help our communities.
…and then, I can write more blog entries.