On containers

Anthony Nocentino (@nocentino) has invited us to write about what we’re doing with containers. It’s kinda his thing. It’s good to have a thing. I got him to speak to the Adelaide User Group about it in February 2020, which turned out to be our last in-person meeting.

(We could switch back to in-person, but the room at Microsoft only holds 15 people under the current restrictions, and I don’t think we’ve ever had attendance that low in the 16 years I’ve been running it.)

Two years ago, I wrote about how I’m running SQL on Linux using docker, and I do find this to be really useful (because I don’t have to spin up a whole VM or make my machine dual-boot as I did back in the 1990s). I’ve also used containers for various Machine Learning tasks. In both situations, it gives me the flexibility to be doing things without having to use my local operating system. In the last couple of years I’ve replaced my main laptop, and the impact on my containers has been really small.

I know I’m not getting as much out of my containers as Anthony is. I don’t have a large testing suite letting me spin up umpteen versions of SQL. But I know that containers will let me do this as and when I need it. I know that I can use containers to spin up things without their having a lasting impact on my local operating system. And that’s really useful.

I encourage you – get used to containers. They’re here to stay.