Doughnut Dilemma: A Lesson in Resource Managers
summary
Uses a doughnut box as a mapping from requests to resources and explains concrete rules: schedules, assignments, constraints like reservation windows and eligibility, and a 12‑donut eating constraint as an example of quota enforcement. Describes DRF (dominant resource fairness) and shows a progression from monolithic/two‑level/shared‑state schedulers with concrete examples: Apache Mesos, Google Omega, Microsoft Apollo, and explains that shared‑state schedulers coordinate placement across resources.
AI-written summary · about
related talks
43mEverything You Ever Wanted to Know About Resource Scheduling, But Were Afraid to Ask
34mPatterns for building Highly Available Cluster Schedulers
21mDemystifying Resource Management in Kubernetes
40mLimits, Requests, QoS... Rethinking What You Thought You Knew About Scheduling in Kubernetes
26mResource as You Wish: Collaborative Reservation and Allocation b...
45mLimits, Requests, QoS, PriorityClasses: Demystifying Scheduling in K8s!
05mResource Roulette: Winning the Kubernetes Allocation Game
10mLightning Talk: Resource Fairness and Utilization for Heterogeneous Batch/M...