San
Jose State University Computer
Engineering
CmpE213 Parallel
Processing / Supercomputers
Instructor:
Dr. Rod Fatoohi Spring
2007
Course overview
Textbook:
Introduction
to Parallel Computing, 2nd ed., 2003, Grama,
et al., ISBN: 0-201-64865-2
Notes
References:
Parallel
Programming in C with MPI and OpenMP, 2004, Quinn,
ISBN: 0-07-282256-2
Parallel
Programming with MPI, 1997, Pacheco, ISBN: 1-55860-339-5.
Exam
& Grading:
20% Project
20% Homework & Lab
Reports
25% Midterm: 7th
Meeting (March 8th) at 1800.
35% Final: 16th
Meeting (May 17th) at 1715.
Exams could be a
mix of problem solving and multiple-choice questions.
No make-ups exams (or
Incomplete grade) except in case of verifiable emergency circumstances.
No laptops during the exams.
•
A+ : > 94
•
A : 90 – 94
•
A- : 85 – 89
•
B+ : 80 – 84
•
B : 75 – 79
•
B- : 70 – 74
•
C+ : 65 – 69
•
C : 60 – 64
•
F : < 60
(0.5
– 0.9) = 1
(0.1
– 0.4) = 0
Academic
integrity statement (from Office of Judicial Affairs)
Your own commitment to learning, as evidenced by your
enrollment at San José
State University
and the University’s Academic Integrity Policy requires you to be honest
in all your academic course work. Faculty are required
to report all infractions to the Office of Judicial Affairs. The policy on
academic integrity can be found at: http://www2.sjsu.edu/senate/S04-12.pdf
Campus policy
in compliance with the Americans with Disabilities Act
“If you need course
adaptations or accommodations because of a disability, or if you need special
arrangements in case the building must be evacuated, please make an appointment
with me as soon as possible, or see me during office hours. Presidential
Directive 97-03 requires that students with disabilities register with DRC to
establish a record of their disability.”
Time
& Location:
Lecture: Thurs 1800 – 1950 DMH 357
Lab: Thurs 2000 – 2250 ENG 206
Office
Hours:
Tuesdays 0900 - 1200 & Thursdays 1700 -
1800, or by appointment only
Office: ENG 273.
Phone: (408) 924-4059.
URL: http://www.engr.sjsu.edu/fatoohi/cmpe213/cmpe213.html
Yahoo Study Group:
normally initiated by a student.
Answering phone calls & checking email are during office hours only.
Prerequisite
Classified
Graduate Standing
Programming in C
Experience with Unix
Attendance:
Highly recommended
Avoid disturbing the class: turn off cell
phones (or put them on vibrate mode), no text messaging during the lecture or
the exams, no entering after 15 minute late, ...
Students are responsible for the lecture,
lab & project presentations
Project: Develop, implement, test, and analyze performance of
a parallel algorithm/application.
·
Typically a group of two students works on a single
project.
·
One report per group: 10 pages (5000 words)
·
Topics will be discussed in the class.
·
Presentation is required for all group members.
·
Proposal (one page per group) Deadline: Meeting #11
·
Report Deadline: Meeting #13 - no late submission
Lab: To provide hands-on experience in developing parallel algorithms
using message passing and shared-memory interfaces.
Course
Description
Parallel processing hardware, software,
and applications. State-of-the-art
description of parallel architectures, communication operations, parallel
programming models, parallel algorithms, and performance analysis.
Learning
Objectives
To provide in-depth description
of high-performance systems and parallel architectures, communication
operations, parallel programming models, parallel algorithms, and performance
modeling.
Outline
This is a
tentative schedule (subject to change):
Meeting Topic Chapter
1 Introduction 1
2 Parallel
architectures 2
3 Interconnection
networks 2
4 Parallel
algorithm design 3
5 Programming
with MPI 6
6 Communication
operations 4
7 Midterm
8 Programming
with Pthread 7
9 Programming
with OpenMP 7
10 Performance modeling 5
11 Matrix
algorithms 8
12 Sorting
algorithms 9
13 Special
topics
14
Project presentations
15 Project
presentations