Kubernetes: Autoscaling is Magic

Aleksandra Malinowska

How many pods do I need to run this service? What machines should I choose? Will there be enough resources in case of a spike in demand? Or maybe I could save that extra core hour at night? What if Im doing it in a suboptimal way? How much memory did it need again?

About Speaker

Aleksandra is a software engineer on Kubernetes autoscaling & zero-ops team at Google.