Advanced Parallel and Distributed Computing
COMP 7330, Fall 2009
Prof. Weikuan Yu (wkyu@auburn.edu)
TR 11:00-12:15pm, RM 304B, Ramsay Hall
Office Hour: Wed 2-3pm, Tel: 334-844-6330
Prerequisite
Prior knowledge in computer architecture and networking is required, which
translates to the following courses at Auburn University, or equivalent
elsewhere.
- COMP 4300, Computer Architecture
- COMP 4320, Introduction to Computer Networks
Course Description
This is a graduate-level course. It focuses on leading system
architecture, high speed interconnects, and programming models that have
been used for parallel and distributed computing environments. This course
will cover fundamental principles, advanced algorithms, and engineering
tradeoffs in building large-scale parallel and distributed computing
systems, as well as the high speed interconnects that bring them all
together. Students will gain an in-depth understanding of research and
development in HPC, and their impact on computational sciences.
Textbooks
- Required: David Culler, J.P. Singh, and Anoop Gupta. Parallel Computer
Architecture: A Hardware/Software Approach. Morgan Kaufmann. ISBN-10:
1558603433
- Optional: William J. Dally and Brian Towles. Principles and Practices of
Interconnection Networks. Morgan Kaufmann. ISBN-10: 0122007514
- Optional: Michael J. Quinn. Parallel Programming in C with MPI and
OpenMP. McGraw Hill Higher Education. ISBN-10: 0071232656
Handouts
- Syllabus
- Schedule
- Chapter 01: Part 1, Part 2
- Chapter 02
- Homework #1 (Due 09/08 11:50pm, a week later for GOP students)
- Chapter 3: Part 1, Part 2
- Chapter 4
- Project Assignment (Homework #2, updated)
- Interconnection Networks:
Reading Document,
Topology,
Switching,
Routing
- Homework #3
- Midterm Review
- Chapter 5: Part 1, Part 2,
State Diagrams
- Homework #4
- Chapter 6
- Chapter 7
- Homework #5
- Chapter 8
- Homework #6
Course Objectives
Students are expected to accomplish the following objectives through this
course:
- Grasp a thorough understanding on the advances of technologies,
system architecture and communication architectdure that propelled the
growth of parallel and distributed computing systems
- Accomplish a good understanding of principles and practices in high
speed interconnects
- Gain an appreciation on the challenges and opportunities faced by
parallel systems, and cloud computing environments
- Obtain an experience on parallel programming and system evaluation
on representative parallel and distributed systems.
Topics
- Advances of technologies and system architectures for parallel and
distributed computing
- Components and organization of high speed interconnects
- Parallel computing algorithms and representative programming models
- Cloud computing and its prevalence
- Convergence of parallel, distributed and cloud computing
- Performance modeling and system evaluation
Grading
- Class participation (10%)
- Course Assignments (30%)
- Midterm exam (30%), in class, 11:00am – 12:15pm, Oct 8th, 2009.
- Final exam (30%), 12:00pm – 2:30pm, Dec 10th, 2009
Office Hours
- Wednesday 2-3pm, or by appointment.
Last updated: August 10, 2009