Practical Container Scheduling: Optimizations, Guarantees, and Trade-Offs at Netflix - Lecture

Practical Container Scheduling: Optimizations, Guarantees, and Trade-Offs at Netflix - Lecture

Linux Foundation via YouTube Direct link

Security

9 of 20

9 of 20

Security

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Practical Container Scheduling: Optimizations, Guarantees, and Trade-Offs at Netflix - Lecture

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Intro
  2. 2 Reactive stream processing: Mantis
  3. 3 Container deployment: Titus
  4. 4 What the cluster needs to support - Heterogeneous mix of workload
  5. 5 Why juggle at all?
  6. 6 Scheduling challenge in large clusters
  7. 7 Our initial goals for a cluster scheduler • Multi goal optimization for task placement . Cluster autoscaling • Extensibility
  8. 8 Multi goal task placement
  9. 9 Security
  10. 10 Capacity guarantees
  11. 11 Fenzo scheduling strategy
  12. 12 Fitness functions we use • CPU, memory, and network in packing
  13. 13 Hard constraints we use • GPU server matching
  14. 14 Soft constraints we use • Specified by individual jobs at submittime • Balance tasks of a job across availability zones
  15. 15 Mixing fitness with soft constraints
  16. 16 Our queues setup
  17. 17 Sizing agent clusters for capacity
  18. 18 Reasoning about allocation failures
  19. 19 What's next?
  20. 20 Questions?

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.