CS2013-final-report

What is the format of the course
0804021 is a 2 credit course with 28 lecture hours and 12 lab hours. Classes typically meet twice per week for
lecture, with lab sessions completed in tutoring labs outside of lecture. Course material is available online. The
course is taught online and face-to-face. Generally there is a combination of lectures, class discussion, case studies,
written term papers, and team research and presentation.
How are students assessed
Students are assessed on a combination of class attendance, group activities, discussion, projects, participation in
parallel programming competition held by enterprises, and a comprehensive final exam.
Course textbooks and materials
• C Lin, L Snyder. Principles of Parallel Programming. USA: Addison-Wesley Publishing Company, 2008.
• B Gaster, L Howes, D Kaeli, P Mistry, and D Schaa. Heterogeneous Computing With Opencl. Morgan
Kaufmann Publishing and Elsevier, 2011.
• A Grama, A Gupra, G Karypis, V Kumar. Introduction to Parallel Computing (2nd ed.). Addison Wesley,
2003.
• J Jeffers, J Reinders. Intel Xeon Phi Coprocessor High-Performance Programming. Morgan Kaufmann
Publishing and Elsevier, 2013.
• T Mattson, B Sanders, B Massingill. Patterns for Parallel Programming. Addison-Wesley Professional, 2004.
Why do you teach the course this way
Being funded by the Project of “Parallel Programming Principle and Practice” from the Ministry of Education
(MOE) of the People’s Republic of China to design a syllabus for senior undergraduate students on parallel and
distributed programming, we serve as the China’s first pilot of restructured courses to integrate topics of
CS2013/TCPP Curriculum Initiative on PDC requirements into the undergraduate curriculum.
The goal of the course is to provide an introduction to parallel computing including parallel computer architectures,
analytical modeling of parallel programs, principles of parallel algorithm design. We will introduce existing
mainstream parallel programming environment and present development situation, which will make the students
understand the basic knowledge of parallel programming. The labs will guide students to use tools to simulate an
optimized parallel program, enable them to master skills to design, code, debug, analysis and optimize some
mainstream parallel software.
The course has a good reputation among students. The students that have taken this course won Best MIC
performance award at “Asia Student Supercomputer Challenge 2013” (http:// www.ascevents.org/13en/index13en.php).
This course has been granted an Early Adopter status by the NSF/TCPP curriculum
committee (http://www.cs.gsu.edu/~tcpp/curriculum/q=list-of-early-adopters-fall-2012.html).
Body of Knowledge coverage
KA Knowledge Unit Topics Covered Hours
PD Parallelism Fundamentals Overview: Parallel computing, architectural demands and trends, goals
of parallelism, communication, coordination
2
PD Parallel Decomposition Independence and partitioning, Data and task decomposition,
Synchronization
1
PD
Communication and
Coordination
Shared Memory, Threads, Message Passing, Consistency 2
- 340 -