Since the Platform Engineering goal is to empower developers to ship apps, it’s fundamental that they at least better understand the basics around application autoscaling. That’s why @Pierre Mavro wrote a concise article https://www.qovery.com/blog/understanding-the-basics-of-application-autoscaling The question that comes to my mind is - how many of you are using a multi-dimensional autoscaler in production? 🤔

Though Multi-dimensional auto scaling is good to have , how many applications really need those ?? And they the engineering teams have enough knowledge to manage this kind of infrastructure ??? Nice summary from an article

That’s a good question TBH.

It's interesting to have if you don't want to manage the control of it. Obviously limits are here and they will be the blocker. But it can help to understand how code big changes cat impact resources, without fearing the downtime in at the very beginning. I personally prefer to chose the horizontal autoscaler and it's a good practice to bench your soft (knowing the limits) before moving it to production. This will avoid all bad autoscaling surprises