Explore optimizing storage assignment through pod scheduling in distributed storage systems like Ceph in this conference talk from KubeCon + CloudNativeCon Europe 2021. Learn how to automate storage allocation using the "WaitForFirstConsumer" volume binding mode and kube-scheduler tuning. Discover the utilization of Topology Spread Constraints to distribute storage pods effectively. Understand the challenges posed by disturbance factors such as CPU-consuming workloads and how to address them. Gain insights into tuning methods for Kubernetes versions 1.17, 1.18, and 1.19 to achieve optimal pod distribution. See a demonstration of how this approach can lead to fault tolerance against full rack failures. Delve into topics including problem statement, basic ideas, port distribution, handling unsatisfiable conditions, and key takeaways for implementing these strategies in your own Kubernetes environments.
Optimizing Storage Assignment via Pod Scheduling Under Disturbance Factors
CNCF [Cloud Native Computing Foundation] via YouTube
Overview
Syllabus
Introduction
Who am I
Agenda
Problem Statement
Kubernetes
Basic idea
Port distribution
When unsatisfiable
Tuning Pod Scheduling
Demonstration
Key takeaways
Taught by
CNCF [Cloud Native Computing Foundation]