Serverless in the Wild - Characterizing and Optimizing the Serverless Workload

Serverless in the Wild - Characterizing and Optimizing the Serverless Workload

USENIX via YouTube Direct link

Fixed Keep-Alive Policy

13 of 21

13 of 21

Fixed Keep-Alive Policy

Class Central Classrooms beta

YouTube playlists curated by Class Central.

Classroom Contents

Serverless in the Wild - Characterizing and Optimizing the Serverless Workload

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

  1. 1 Intro
  2. 2 What is Serverless?
  3. 3 So what are people doing with Faas?
  4. 4 If you are a cloud provider
  5. 5 Stepping Back: Characterizing the Workload
  6. 6 Invocations per Application
  7. 7 Apps are highly heterogeneous
  8. 8 What about memory?
  9. 9 Function Execution Duration
  10. 10 Key Takeaways
  11. 11 Cold Starts and Resource Wastage
  12. 12 What do serverless providers do?
  13. 13 Fixed Keep-Alive Policy
  14. 14 Fixed Keep-Alive Won't Fit All
  15. 15 Fixed Keep-Alive Is Wasteful
  16. 16 A Histogram Policy To Learn Idle Tinil
  17. 17 The Hybrid Histogram Policy
  18. 18 More Optimal Pareto Frontier 130
  19. 19 Implemented in OpenWhisk
  20. 20 Simulation
  21. 21 Closing the loop

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.