Databases and “.Net Rocks”
8 January 2015, by David Simons
I love databases.
I know, it sounds geeky, but I really do find databases fascinating. They idea of storing data has such intuitive sense, but in practise, it requires such performance optimisation that getting to grips with a new flavour of database can be daunting. I remember the antipatterns employed when I first saw SQL in my teens (“An unlimited number of items? Let’s make three letter codes delimited by 999!”), but now normal forms flow naturally from my fingers.
I’ve been lucky enough to talk a lot about this topic recently, and specifically, I wanted to remind people that there were more options than just SQL out there! That’s not to say I’m one of these developers that things that SQL isn’t fit for purpose – any system that has been used so comprehensively for so long has something going for it – but I’ve overheard a lot of conversations that assume that SQL is going to be plonked at the bottom of a tech stack without much care and attention. That’s why I’m telling anyone who will listen there is other stuff out there.
If all you care about is a series of simple numbers recorded at certain times, then why go through the pain of setting up a SQL instance when you can use a key/value store, or even farm your data out to one of the many database as a service companies. On the flip side, sometimes SQL isn’t complex enough – and you’re dealing with data that would be better served by a graph database.
I was invited to talk (or maybe rant) on this topic with the team behind .Net Rocks, one of my favourite technical podcasts. You can listen to my discussion (as well as over 1,000 previous episodes) on their website – where we manage to discuss over ten different databases in an hour.
I hope you enjoy listening to it as much as I enjoyed recording it.