Completed
SLearn: A Novel Approach for Learning Runtime Properties
Class Central Classrooms beta
YouTube playlists curated by Class Central.
Classroom Contents
A Case for Task Sampling Based Learning for Cluster Job Scheduling
Automatically move to the next video in the Classroom when playback concludes
- 1 Authors Introduction
- 2 Challenges in Cluster Scheduling
- 3 Learning Runtime Properties for Cluster Scheduling
- 4 Widely Used Approach for Learning: History-based Learning
- 5 History-based Learning: Assumptions and Reality
- 6 Poor Performance of the State-of-the-Art History-based Predictor
- 7 SLearn: A Novel Approach for Learning Runtime Properties
- 8 Learning in Time History vs Learning in Space SLearn
- 9 Comparing Prediction Accuracy: Large Scale Trace-based Analysis
- 10 Comparing Coefficients of Variations CoVs across Space and Time
- 11 Varying the History Length in CoV comparison
- 12 Comparing Prediction Overhead: Simulation and Testbed Experiments Using GS
- 13 SLearn's Implementation and Design
- 14 Baselines and Experimental Setup
- 15 Simulation and Testbed Experimental Results
- 16 SLearn for DAG and Future Work
- 17 SLearn Summary
