A collection of articles I've written. Software architecture. Optimization. Random thoughts. AI.
Do you like reviewing someone else's AI-code? It doesn't feel the same, right? This is something that's been on my mind recently - is the (human) code reviews about to be killed by AI? What happens when we let an AI write our code?
Have you ever wanted to reduce the disk space used by an Azure SQL database?
You will quickly find out that the general advice on the Internet is
Don't do it!
. But how bad is it? Read on for a report on how I reclaimed 100s of GB and
saved 2500+ USD from our monthly Azure SQL cost with zero downsides!
One of the classic bugs is the memory leak. Even garbage collected languages like C# can leak memory, even if all the code is safe/managed. Sometimes such issues are easy to find and fix. Other times its difficult, for example when the memory leak only happens in production, only sometimes, in a containerized .NET service running in Kubernetes. This difficult case is the one I will cover in this post.
It started with this quote: We hit 100% DTU on user DB again yesterday
.
Follow along into a deep dive into optimizing the unindexable
Msisdn LIKE '%ddd%' SQL query.
It is not easy to be a AI DBA
Do you also have the feeling AI advice often tricks you as much as it helps? I challenged a colleague who specified a fill factor of 80 of when creating a new table with a non-clustered index (in Azure SQL/SQL Server) and we asked the AI for help.
Redis is arguably the most well regarded tech of all. But many times you do not need Redis. Here are 3 examples from my last 3 jobs.
During a AI Hackathon at work I trialed a AI-powered code review plugin. Did it work?
What is more fitting as a first post than an overview of the tech used to build this site?