Boston University | Center for Computational Science
HomeNews and EventsResearchEducationPeopleSeminarsFacilitiesContact Us

Parallel Computing: Models, Languages & Architectures

CAS CS551 - Spring Semester (alternate years)
A. Bestavros

Over the last few years, the emergence of radically different parallel computer architectures has prompted the development of various programming models such as data flow, shared memory, message passing, systolic, and data parallel, all of which are architecture-dependent models. The rapidly changing grounds of parallel architectures, however, suggest that architecture-independent models such as Unity be used, thus leaving the task of tailoring a given program to a particular architecture to an optimizing compiler.

This course aims at exploring these two alternatives by introducing a palette of programming models and contrasting their suitability for different architectures and applications. The material covered will encompass topics in parallel computer architectures, parallel programming models, and languages. Appropriate examples for existing or proposed parallel architectures will be surveyed. Alongside, students will have the opportunity to gain hands-on experience with a host of architecture dependent/independent parallel programming languages, such as Unity, Id, Linda, C*, Fortran 90, and MPL.

Co-Requisite Laboratory/Seminar

Students from the advanced parallel computing courses meet together as a group in the weekly Laboratory/Seminar. The Seminar provides a vehicle for interdisciplinary presentations by students of their project work in the courses and serves as a forum for instruction in advanced computational techniques. The meetings are held in the advanced workstation laboratory specially designed for these courses to provide for interactive participation.

Page last updated 06/05/2006. Please send comments to Cheryl Endicott

copyright © 2006, Center for Computational Science | Boston University , MA, 02215