I watched the following talks:
- Hadoop Primer for SQL Server DBA by Konstantin Korobkov at SQLBITs XIV (The sound quality for this one isn't great.)
- Python and R for SQL and Business Intelligence Professionals with Jennifer Stirrup at SQLBits XIV.
- Shiny: dashboards in R by Steff Locke (@SteffLocke) SQLBITS XIV
- "How to Build a Virtual Test Lab for SQL Server" with Ed Leighton-Dick & David Klee at SQLPASS Summit 2015
- "Analyzing your ETL Solution with PowerShell" with André Kamman at SQL PASS Summit 2015
- "Change Data Capture Case Study and Checklist" with Uwe Ricken at SQL PASS Summit 2015
The most stand-out moments from last week's efforts include:
- I altered a SQL Server partitioning configuration that I had set up a number of years ago. Even though I implemented partitioning as a "Plan B" for this particular application, it has worked surprisingly well. I haven't had to touch the configuration since I implemented it. I wouldn't have to touch the partition configuration now, but application has survived well past it's initially estimated retirement date.
- I improved my Pester skills. I've got a lot of infrastructure and configuration tests running now. The weak point, I think, is the overall organization of what I've built. Looking at what I've done, it seems haphazard to me, but I am not entirely sure what "better" should look like. Everyone seems to have their own manner of breaking out or categorizing things via file/Describe/Context/It and their own way of looking through items to be tested. The sqlcollective/dbatools folks intend to release a set of best practice recommendations implemented as Pester tests this spring. I intend to raid them for ideas on how to organize dozens of tests of various sorts. I'm also sure that there is some overlap between what they intend to roll out and what I've been doing. I may need to retire some bespoke code in favor of community-supported code.
- I installed the latest Pester on a couple of my workstations, updating from 3.4.x or so to 4.1.x. I have seen one or two differences. I haven't bothered to read about the improvements that come with the new version yet (though I do like -Show very much).
- I removed several dependencies on external code from a code base that I maintain.
- I updated a toy solution of mine, though the SSIS project in the solution proved to be a problem. The take-away is: If your copy of Visual Studio 2017 (which you *know* has SSDT installed ) doesn't seem to support BI projects (SSRS, SSAS, or SSIS), then you probably don't have the latest version of SSDT installed. VS doesn't seem to update SSDT bits as part of the usual "update extensions & tools" process.
- I managed to get through a debugging session in VSCode without it crashing on me. This is a first for me. Either VS Code is getting better or I've learned something about how to use the debugging modes that VS Code provides.
I'd also like to promote the next Philly-area SQL Saturday a tiny bit. SQL Saturday 714 will be held at Montgomery County Community College on April 21, 2018. I skipped SQL Saturday last year, so I haven't been since the event was held at Microsoft's facility in Malvern and I'm curious to see how it will work at this particular venue.