Professor: Tia Newhall
Semester: Spring 2008
Time: 11:20-12:35, TR
Location: Science Center 246

Office hours: 2-4 W and by appointment

Course Description

This course covers a broad range of topics related to distributed systems. Distributed systems consist of a set of PCs or workstations connected by a network, that run special software that allows for transparent sharing of the distributed computing resources and data.

We will investigate ways in which computer resources connected by a network can be used and shared. This includes using them for parallel computing (ex. cluster and grid systems), for resource and file sharing (ex. grid systems, BitTorrent), for data and communication (ex. WWW, digital libraries, IM, email), and for availability (ex. when a Google server goes down, how does Google completely hide that fact from you?).

Class will be run seminar-style; each week students will read research papers that we will discuss in class. We will read papers that cover the theory of distributed systems as well as the implementation of systems to support distributed computing.

During the first part of the course, we will examine different parallel and distributed programming paradigms and you will get some practice writing programs using these different paradigms. In distributed systems, parallel computing is used in cluster and grid environments, and distributed computing is used in client-server systems such as the WWW and peer-to-peer systems.

During the second part of the course, students will propose and carry out a semester-long research project related to distributed computing. The department's gigabit cluster, as well as the CS lab machines, are available for course projects.

Previous course work in Operating Systems, Networking, Databases, or Architecture is helpful, but not necessary for taking this course.

Course Goals

• Analyze and critically discuss research papers both in writing and in class
• Formulate and evaluate a hypothesis by proposing, implementing and testing a project
• Relate one's project to prior research via a review of related literature
• Write a coherent, complete paper describing and evaluating a project
• Orally present a clear and accessible summary of a research work
• Understand the fundamental questions addressed in the class and analyze different solutions to these questions

---

Grading

• Class Participation (40%)

  There are four components of your class participation grade:
  1. Participation in class discussion of research papers
     You are expected to come to class prepared to discuss the reading assignment for the day. See my Tips for Reading Research Papers.

  2. Reaction Notes
     Before 1am of most class days, you will post your reaction notes to the day's reading using handin85. See Reaction Notes, for more information about how to structure your reaction notes.

     Reaction notes are graded on this scale (note: in computing final grades, the real difference between a 0 and 1 is much greater than the real difference between a 1 and 2):

     • 0: didn't submit
     • 1: submitted, but not complete (this usually means that your discussion is at too high a level, and that I want more explaination)
     • 2: submitted, complete

  3. Presentation of Research Paper(s) on sub-topic
     Each of you will give an in-class presentation on a sub-topic that we will cover. In addition to reading the weekly research papers, you should read additional related papers. Based on your reading you will prepare a ~50 minute presentation using html, pdf, powerpoint slides. By the Friday prior to your presentation, you should meet with me with an outline of your talk, and you should meet with me again at least by Monday (Wednesday) prior to your Tuesday (Thursday) talk to go over your slides. See my Oral Presentation Guide for information on how to structure a research presentation.

     Beginning with the week 3 topics, you can pick any topic on which you will give a presentation.

  4. Evaluation After each class meeting you will fill out a self-evaluation and summary form. Also, on days when we have class presentations, you will provide written feedback to other students on their presentations.

• Programming and Written Assignments (10%)

  During the first half of the semester you will be assigned a few small programming assignments to give you some practice writing parallel programs. In addition, one or more short written assignments may be assigned.

• Course Project (50%)

  You will work on a semester long course project related to distributed or cluster computing. Components of your grade include an annotated bibliography, a written project proposal, a written and oral midterm progress report, and a final written report, oral presentation, and demo of your project.
  Project Components and Details

---

Reading List

There is no textbook for this class. Instead, we will read papers from the Reading List.

---

Class Schedule

WEEK	DAY	ANNOUNCEMENTS	TOPIC	READING & ASSIGNMENTS
1	Jan 22		Distributed Systems Overview	week 1 readings
2	Jan 29		Communication and System Design	week 2 readings homework 1
3	Feb 05		Message Passing, RPC	week 3 readings homework 2
4	Feb 12	Course Project Information	Time, Event Ordering, Distributed State	week 4 readings homework 3 Course Project Assignment
5	Feb 19		Distributed File Systems	week 5 readings
6	Feb 26		Grid Computing	week 6 readings
7	Mar 04	Project Proposal due (Mar 06)	Project Work Week	week 7 assignments
	Mar 11	Spring Break
8	Mar 18		Peer-to-Peer Systems	week 8 readings
9	Mar 25		Security, Distributed Shared Memory	week 9 readings
10	Apr 01	Project Progress Report due (Apr 03)	Fault Tolerance	week 10 readings Project Progress Report
11	Apr 08		MapReduce	week 11 readings
12	Apr 15	Project Progress Report due (Apr 17)	Cluster Computing	week 12 readings Project Progress Report
13	Apr 22		Process Migration, Scheduling	week 13 readings
14	Apr 29		Individual Project Group Meetings and Final Project Presentations	week 14 scedule final project report, presentation, demo
15	May 06	Final Project Report due by noon (May 09)	Finals	no reading

---

Resources

Distributed Systems, Parallel systems, and Cluster links

• Some Cluster Computing Links
• IEEE Distributed Systems Online
• The Top 500 Supercomputers
• ParaScope IEEE Listing of Parallel Computing Sites
• HPC Wire
• Introduction to Parallel Computing from Livermore Computing

Reading, Writing, Presentation Advice

• Tips for Reading Research Papers
• Research and Writing Guide This is written as a CS honors thesis guide, but is applicable to the type of research and writing that you will do for your course project. Look at the "Getting Started" part for links to starting points for finding related work to your project.
• Oral Presentation Guide Guidelines for preparing a research talk, and links to other oral presentation advice.

Parallel and Distributed programming links

• Links to Parallel and Network Programming Resources mpi, openmp, posix threads, socket programming, ...
• My CS and Unix Documentation make, tar, cvs, rcs, debugging tools, editors, email, programming guides and tutorials, linux and parallel computing links
• CS Department's Unix Help Pages
• CS Project Etiquette Tools and Guidelines for using shared CS resources to run long-running, intensive applications.
• Tools for running experiments and collecting performance measurements