CSE 4143/6163

DESIGNING PARALLEL ALGORITHMS

Prof. Ioana Banicescu.

HOURS:

CLASS: Tue/Thu 2:00 - 3:15p.m, Butler 104.

OFFICE: 320 Butler Hall, Tue/Thu 3:15-5:00. Other times by appointment only. Email: ioana@cs.msstate.edu.

325-7508.

TEXT BOOK:

Vipin Kumar, Ananth Grama, Anshul Gupta, George Karypis, Introduction to Parallel Computing, Design and Analysis of Algorithms, The Benjamin/Cummings Publishing Co., Inc. 1994.

RECOMMENDED READINGS FROM:

1. Ian Foster, Designing and Building Parallel Programs, Addison-Wesley Puplishing Co., 1995. Also available at .

2. W. Gropp, E. Lusk and A. Skjellum, Using MPI, Portable Parallel Programming with the Message-Passing Interface, The MIT Press, Cambridge MA, 1994.

3. Peter Pacheco, Parallel Programming with MPI, Morgan Kaufmann Publishers, 1997.

4. M. Snir, S. Otto, S. Huss-Lederman, D. Walker, J. Dongarra, MPI The Complete Reference, The MIT Press, Cambridge MA, 1996.

PREREQUISITES:

The required prerequisites are listed in the official university bulletin. Graduate level courses in algorithms, parallel algorithms, operating systems, computer architecture are desirable. Knowledge of Unix, C and Fortran, enough to do simple programming assignments are recommended.

OBJECTIVES:

The goal of this course is to familiarize the studends with efficient algorithms for parallel and distributed systems. The course will be focused on evaluating the algorithms theoretically as well as implementing them on a network of workstations and cluster of SMP UltraMSPARC at the Engineering Research Center(ERC).

COURSE REQUIREMENTS:

This course examines a variety of issues in the design of parallel algorithms such as: what algorithms and data structures should be used when programming parallel computers; how can the quality of these algorithms be analyzed; what are the factors that influence the performance of algorithms on parallel machines. The students will study features of some parallel programming paradigms in the abstract and will learn important aspects in algorithm impementation by completing several programming projects.

There will be a midterm and a final exam, several homeworks and programming projects.

GRADING: Homework and Programming Assignments (20%-grads, 40%-ugrads), Project (20%-grads only), Midterm (20%), Final (40%).

ANNOUNCEMENTS:

Students are responsible for checking the announcements regularly. For important announcements, click here.

COURSE OUTLINE:

The course will cover material from the following topics:

Fundamentals: Introduction to parallel and distributed computing, models of parallel computers, parallel programming models.

Design methodology for parallel algorithms.

Basic communication operations; introduction to MPI.

Performance evalutation of parallel systems: Performance metrics, execution time, speedup, efficiency, cost, scalability, isoefficiency and cost effectiveness.

Sorting algorithms: Sorting on parallel computers, sorting networks, bubble sort, quick sort and other sorting algorithms.

Graph algorithms: Definitions and representation, minimum spanning tree, single source shortest paths, all pairs shortest paths, transitive closure, connected components, algorithms for sparse graphs.

Dense matrix algorithms: Mapping matrices on processors, matrix transposition, matrix vector multiplication, matrix multiplication, solving systems of linear equations.

ACADEMIC HONESTY:

Students are expected to follow the CS department's policy on academic honesty with regards to homework, projects and exams.