Shop Talk: 2020-10-19

The Recording

The Panelists

  • Kevin Feasel
  • Mala Mahadevan

Notes: Questions and Topics

Announcements

We had two announcements. First, TriPASS board elections are coming up in mid-November. We have three positions up for election this year: President, VP of Marketing, and Treasurer. All TriPASS members in good standing of eligible to run, and the election will run from November 19th through December 3rd, for terms beginning February 23, 2021 through February 26, 2023. If you’re interested, e-mail (shoptalk at tripass dot org will work) and I can provide more details as needed.

Second, Mala brought up the Azure SQL Championship, so check that out.

Date Buckets in Azure SQL Edge

Our first topic for the night was a function available only in Azure SQL Edge: DATE_BUCKET(). I walked through a sample of some of the code I plan to show off during my PASS Summit pre-con, and I’m glad I did this because I was not very happy with the delivery. The concept feels a little backwards to me, as “date bucket” sounds like it should be numeric: bucket 1, bucket 2, bucket 3. Instead, it’s the begin date for the bucket itself: 2020-09-01, 2020-09-15, 2020-09-29.

I certainly believe it’s useful and would like to see it in SQL Server and Azure SQL Database, but I’ll need to work on my examples to make sure they’re clearer.

Thoughts on Graph Databases

The other key topic for the night was a discussion of graph databases. We recently recorded an episode of the SQL Data Partners podcast with Mala as the guest and the two of us had some detailed thoughts on the utility of graph databases. We took some time tonight to flesh out those thoughts a bit further.

Mala shared her thoughts and included a useful video on graph databases, including their pros and cons.

The short version of it for me is that graph concepts are extremely useful; graph databases are not, unless your company’s key differentiator requires solving graph-style problems with large amounts of data in near real time. An example I like to give for this is logistics, in which you need to know where people and things are and where they need to go in the most efficient manner. If you’re developing software for logistics companies to use to solve this problem, then a graph database might be the best option. Check out the show (as well as the upcoming podcast episode, #208) for the rest of my thoughts on the topic.

Mala’s Books & More Kiosk

We’ve upgraded Mala’s Book Corner to include a lot more. This time around, she has one blog post and three podcasts for us.

Leave a Reply

Your email address will not be published.