EECS 122, Lecture 26
Another Look at Sharing
Problem Set #5 (the last)
Why QoS Anyhow?
Basic Service Model
Expressing Reservations
Traffic Regulation and Policing
Common Traffic Descriptors
Average Rate Descriptor Window
Linear Bounded Arrival Processes (LBAPs)
Leaky Bucket Regulators
Leaky Bucket Operation
Properties of Leaky Bucket
Example
Constructing End-to-End QoS
What Can be Scheduled?
Scheduling Disciplines
Requirements of a Scheduler
Ease of Implementation
Protection and Fairness
Max-Min Fairness Criteria
Water-Filling Analogy
Max-Min Weighted Fair Share
Max-Min Example
Weighted Max-Min Example
Performance Bounds
Common Performance Bounds
Common Performance Bounds-2
Common Performance Bounds-3
Choices in Scheduler Design
Priority Scheduling
Work Conserving or Not
Non-Work-Conserving Approach
Eligibility Time Computations
Perspective
Aggregation
Service Order
Email: kfall@cs.berkeley.edu
Home Page: http://www-inst.eecs.berkeley.edu/~ee122