Self-Learning Disk Scheduling

Free registration required

Executive Summary

The performance of disk I/O schedulers is affected by many factors such as workloads, file systems, and disk systems. Disk scheduling performance can be improved by tuning scheduler parameters such as the length of read timers. Scheduler performance tuning is mostly done manually. To automate this process, the authors propose four self-learning disk scheduling schemes: Change-sensing Round-Robin, Feedback Learning, Per-request Learning, and Two-layer Learning. Experiments show that the novel Two-layer Learning Scheme performs best. It integrates the workload-level and request-level learning algorithms. It employs feedback learning techniques to analyze workloads, change scheduling policy, and tune scheduling parameters automatically.

  • Format: PDF
  • Size: 3092.48 KB