Kubernetes resource management
Chapter 3,Getting Started with Kubernetes mentioned that Kubernetes has a scheduler that manages Kubernetes node and then determines where to deploy a pod. When node has enough resources such as CPU and memory, Kubernetes administrator can feel free to deploy an application. However, once it reaches its resource limit, the Kubernetes scheduler behaves different based on its configuration. Therefore, the Kubernetes administrator has to understand how to configure and utilize machine resources.
Resource Quality of Service
Kubernetes has the concept of Resource QoS (Quality of Service), which helps an administrator to assign and manage pods by different priorities. Based on the pod's setting, Kubernetes classifies each pod as:
- Guaranteed pod
- Burstable pod
- BestEffort pod
The priority would be Guaranteed > Burstable > BestEffort, which means if the BestEffort pod and the Guaranteed pod exist in the same node, then when one of the pods consumes memory and to causes...
