Internet Algorithms

Spring 2004

Google, Napster, Seti@Home -- it is intriguing how Internet-scale distributed systems work, and that it is often difficult to make them work well. During the Internet Algorithms course, you will learn about several algorithmic problems that arise in large-scale networks, and attempts at solving these problems. We will cover: web search engines, peer-to-peer systems for file sharing, grid computing platforms for massive computational problems. The course will be fairly rigorous (i.e., theorems will be proven).

Course objectives:
*

teach students about several algorithmic problems that arise in large-scale networks, and attempts at solving these problems

* introduce students to current and "hot" research topics
* jump-start research projects that each student will begin to pursue
Prerequisites:
 

mathematical maturity and experience in algorithm analysis is encouraged; course is open to graduate students; undergraduate student interested in taking the course should ask the instructor; a graduate algorithms course, or equivalent background, is required

Readings:
 

there will be no textbook for the course; you will be expected to read the papers covered during the course (example Google's PageRank); see here for a detailed list of papers covered in class

Grade components:
* homeworks there will be a few homeworks assigned during the semester, you can discuss ideas with other students but the solutions must be written down entirely on your own
* presentation you will need to present one sophisticated course-related paper of your choice by the end of the semester; each student should present a different paper
* final report a survey of a few papers about one area will be due towards the end of the semester; the survey should include nontrivial observations of your own, such as posing an interesting new problem, providing an original solution to a problem, or clarifying proofs from the papers; no shared reports
* class participation you are expected to attend the classes and participate in the discussions
* no final examination  
Makeup policy:
 

if you miss any homework deadline, do not present during the slot assigned to you, or fail to deliver the final report before the deadline, for any excused reason, then you will be assigned extra tasks to perform

Links:
  see similar courses offered at ETH, Iowa State, UMass, and UTexas
Course information:
  CS 691-002 Internet Algorithms, Spring 2004
  Instructor: Grzegorz Malewicz
  Lecture: HO 108, Tue/Thu 12:30-1:45
  Office hours: Wed 10:00-12:00, also by appointment