Рет қаралды 45
Kubernetes workload scheduling is built on simple mechanics like requests, limits and priorities, but those mechanics can interact in ways that you might not expect. Fortunately, if you understand the key details, you can avoid most of the nasty surprises.
Join Joe Thompson and Kamesh Pemmaraju as they discuss the key mechanisms Kubernetes provides for managing resource allocation and consumption of workloads, including:
👉 Basics of requests and limits.
👉 Quality of service: How the Kubernetes scheduler implements it and how it assigns a particular Quality of service category to your applications.
👉 Priority and how it interacts with Quality of service classification.
👉 Issues that can arise from common misunderstandings or non-optimal combinations of workload resource configuration, priority, and other options such as affinity.