General Information






Course Description


The network calculus has evolved into an elegant framework for analytical performance evaluation of data networks. It provides methods to determine resource requirements of traffic flows in a network using an envelope description for arrivals and service. More recently, stochastic extensions of the network calculus, referred to as Stochastic Network Calculus have resulted in the discovery of new scaling laws of network delays, and an end-to-end analysis of a network with heavy-tailed traffic. This course discusses progress and discoveries on delay and backlog analysis over the past 20 years, that have led to the Stochastic Network Calculus methododology.

Lecture Schedule and Readings:


Week Topics Material Reading
Week 1
(Sep 12-16)
Packet Switches
Introduction to Deterministic Analysis
PPT
PPT, PPT

[LectureNotes, Chp. 1.1-1.4]
Week 2
(Sep 19-23)
Traffic Envelopes and Regulators
[LectureNotes, Chp. 2.3-2.5]
Week 3
(Sep 26-30)
Introduction to Statistical Multiplexing
Bufferless Multiplexer
PDF   (PPT)
PDF   (PPT)
[LectureNotes, Chp. 5.1]
[LectureNotes, Chp. 5.2]
Week 4
(Oct 3-7)
Statistical Envelopes [LectureNotes, Chp. 5.3-5.4], [Boorstyn00]
Week 5
(Oct 10-14)
Lectures will be rescheduled
Week 6
(Oct 17-21)
Effective Bandwidth
Exponentially Bounded Burstiness
[LectureNotes, Chp. 5.8], [Weiss95]
[LectureNotes, Chp. 5.6], [Yaron93]
Week 7
(Oct 24-28)
Statistical Sample Path Envelopes [LectureNotes, Chp. 6.1-6.4],
Week 8
(Oct 31-Nov 4)
Min-Plus Algebra
Service Curves
[LectureNotes, Chp. 1.5],
[LectureNotes, Chp. 2.1-2.2],
Week 9
(Nov 7-11)
Performance Bounds (deterministic) [LectureNotes, Chp. 3.1-3.3],
Week 10
(Nov 14-18)
Statistical Service Curves [Ciucu06], [Ghiassi11]
Week 11
(Nov 21-25)
Statistical Performance Bounds [Ghiassi11]
Week 12
(Nov 28-Dec 2)
Scheduling and Delay Bounds (PPT) [Boorstyn00] [Liebeherr10]



Problem Sets





Reading List


[LBT] refers to J.Y. Le Boudec, P. Thiran: Network Calculus, Version May 10, 2004.
         Available online at http://ica1www.epfl.ch/PS_files/netCalBookv4.pdf
.
[SNC] refers to Y. Jiang, Y. Liu: Stochastic Network Calculus, Springer 2008.
         Available online at http://www.springerlink.com/content/978-1-84800-126-8#section=43531&page=3&locus=27
(Note: To access this book, you must be on the UofT network or login via the UofT portal)
.

[LectureNotes] refers to class notes written by the instructor (in draft form). Class notes are posted on Blackboard.

DOI refers to the Digital Object Identifer where online versions of the papers can be accessed, e.g., IEEE XPlore. When off-campus, you need to login to the library with your UTORid, to gain access to the papers.

 

 

 

 

 

 

 

 

 

General Info

Lectures

Problem Sets

Reading List