Skip to main content

CSC 579 Introduction to Computer Performance Modeling

3 Credit Hours

(also offered asĀ ECE 579)

This course focuses on the mathematical techniques and procedures required in performance modeling of computer and communication systems. The major mathematical elements of applied probability, stochastic processes, especially Markov chains, and elementary queuing theory, including an introduction to queuing networks, will be discussed. Simulation techniques will also be covered.

Prerequisite

Undergraduate course in probability theory or consent of instructor. Good working knowledge of a high-level programming language such as C, C++, or Java.

Course Objectives

Students will learn mathematical theory and practices that form the foundations of techniques for the performance analysis of computer and communication/network systems.

Analytical tools (e.g., Markov chains) for performance analysis goes far beyond the typical computer network scenarios. While this course is mainly based on the traditional queueing system as a primary application (still a mother of all theory!), this semester, I plan to give a brief account of more modern applications which can still be accessible via Markov chains. Examples include sampling, algorithms on graphs, stochastic optimization and machine learning theory, to name a few.

Course Outline

  • Review of probability theory and random variables, conditional expectation
  • Poisson process
  • Stochastic processes and Markov chains
  • Introduction to Queueing theory and examples
  • Markov chains for computing, sampling, optimizations
  • Introduction to Markov chains Monte Carlo (MCMC)

Course Requirements

Homework 15% (conventional problem solving)
Simulation Project15%
Midterm exam20%
Final exam35%

Textbooks

(not required to buy)

Updated: 4/24/2025