Session: Autoscaling Pods and Nodes in Kubernetes: Where to Start?

Do you want to stop manually scaling Kubernetes pods and manually requesting new hardware every time your service usage spikes?

Do you wonder what types of autoscaler you should be using, or how to tune the parameters for your particular service workload?

Join Katie Wassell, of the IBM Watson Performance Engineering Team, for an overview of the different autoscaling options available in Kubernetes 1.16, including Cluster Autoscaling, Horizontal Pod Autoscaling on CPU and on Custom Metrics, and Vertical Pod Autoscaling. Katie will go over the most common use cases she’s seen for each of the autoscalers, as well as strategies that her team uses to visualize and tune autoscaling artifacts based on production workloads.